Progress in Digital and Physical Manufacturing: Proceedings of ProDPM’21 3031338898, 9783031338892

Table of contents :
Preface
Conference Committees
Invited Lectures
Conference Awards
Contents
Keynote and Workshop
Integration of Additive Manufacturing in Production Systems
1 Introduction
2 A Key Actor: France Additive
3 A Key Player: Additive Manufacturing
4 Key Issues of Additive Manufacturing in Relation with Industrial Production
4.1 Innovative Creation of Innovative Products
4.2 4–2 Spare Parts
4.3 4–3 Digital Logistics and Distributed Production
4.4 4–4 Direct Production
4.5 4–5 Production on Demand – Zero Stock
5 Conclusions
References
How Virtual and Augmented Reality Can Boost Manufacturing
1 Introduction
2 Literature Review
3 Technology Overview
3.1 An Overview About Virtual Reality
3.2 An Overview About Augmented Reality
4 Industrial Application
4.1 Five Main Areas in Industrial Application
4.2 Suggestions on How to Implement Augmented or Virtual Reality Projects
4.3 Example of Tools Developed Using Augmented or Virtual Reality
5 Conclusions
References
Advanced Manufacturing Technologies
CAD/CAM Process Chain for Hybrid Additive Manufacturing
1 Introduction
2 Materials and Methods
2.1 CAD/CAM Process Chain
2.2 Experimental Validation
3 Results and Discussion
4 Conclusions
References
On the Quality of Electron Beam Melted Thin-Walled Parts with Curved Surfaces
1 Introduction
2 Material and Methods
2.1 Overhang Characterization and Production
2.2 Analysis of Thin-Wall Curved Surfaces
3 Results and Discussion
4 Conclusions
References
Computational Origami Based Design in 4D Printing
1 Introduction
2 Origami CAD Software
3 Materials and Methodology
3.1 Materials and Software
3.2 Fabrication
3.3 Shape Recovery Test for the Origami Structure
4 Results and Discussion
5 Conclusion
References
Novel Extrusion Based Co-axial Printing Head for Tissue Engineering
1 Introduction
2 Theoretical Design of Co-axial Nozzle Geometry
2.1 Negative-Protrusion Compression Co-axial Nozzle
2.2 CFD Simulations
3 Co-axial Nozzle Design and Optimization
3.1 Co-axial Nozzle Modelling
3.2 Evaluation of CFD Simulations
4 Conclusions
References
Localization and Control of a Mobile Robot for Additive Manufacturing
1 Introduction
2 Related Work
3 System Design
3.1 Kinematic Model
3.2 Localization Module
3.3 Control Module
3.4 Communication Module
4 Experimental Work
4.1 Accuracy Assessment
4.2 Precision Assessment
5 Conclusion and Future Work
References
Development of a Large Size 3D Delta Printer for Advanced Polymers
1 Introduction
2 Printer Developed
3 RepRap Firmware and Controller
4 Calibration
5 Z-probe Sensor
5.1 Infrared vs. Strain Gauge-Based Sensor
5.2 Load Cell Signal Conditioning
6 Conclusions
References
Modelling the Material Removal Process of Turbulent Jet Electrochemical Machining of Copper
1 Introduction
2 Methodology
2.1 Simulations with 3D Quarter-Body and 2D Axisymmetric Geometries
2.2 Calculations of Material Removal
3 Results
4 Conclusion
References
Size Matters, Designing for Larger AM Products!
1 Introduction
2 Background
2.1 Hardware
2.2 Software
2.3 Case Studies
2.4 Value in the Market
3 Discussions
4 Conclusion
References
Design and Green Manufacturing - CAD and 3D Data Acquisition Technologies
Rapid Product Development: A Decisionmaking Matrix for the Manufacturing of Injection Mould Inserts for Small Batch Production
1 Introduction
2 Methodology
2.1 Part Characteristics
2.2 Insert Characteristics
3 Results and Discussion
3.1 Discussion
4 Conclusion
References
A Methodical Approach to Product Development in 4D Printing Using Smart Materials
1 Introduction
2 Literature Review
3 Methodical Approach
3.1 Analysis of the Task
3.2 Solution Search
3.3 Conceptional Design
3.4 Additive Manufacturing of the Component
4 Application Example
4.1 Analysis of the Task
4.2 Design for 4D-Printing
4.3 Additive Manufacturing of the Component
4.4 Deployment of the Complete System
5 Conclusion
References
Product Design for the Circular Economy: A Design Process for Footwear
1 Introduction
2 Background on the Fashion Industry and Footwear Sector
3 Circular Economy (CE)
4 Circular Product Design (CPD)
4.1 Circular Product Design Strategies
5 Circular Design Process: A Case Study
5.1 Starting Considerations on Design Process
5.2 Understand
5.3 Define
5.4 Make
5.5 Release
6 Concluding Reflection
References
Design of Playful-Pedagogical Objects for Learning and Development of Preschool-Aged Children with Autism Spectrum Disorders (ASD)
1 Introduction
1.1 Disability
1.2 Design as the Answer
1.3 Digital Fabrication
2 Methodology
3 Results
4 Conclusions
References
3D Pine Tree Geometry Design in Forest Fire Environments
1 Introduction
2 Numerical Model and Methodology
3 Results and Discussion
4 Conclusion
References
Recycled Reinforced PLA as Ecodesign Solution for Customized Prostheses
1 Introduction
1.1 Customization with Additive Manufacturing
1.2 Additive Manufacturing Environmental Sustainability Through Recycling
2 Materials and Methods
2.1 Recycled Reinforced PLA for Additive Manufacturing
2.2 Numerical Model
3 Results and Discussion
3.1 Biomechanical System Simulation
3.2 Socket-Type Prostheses Simulation
3.3 Discussion
4 Summary and Conclusions
References
Digital Manufacturing and Simulation Systems
Virtualization and Optimization of Processes in Industry 4.0
1 Introduction
2 Case Study
2.1 Description of the Physical Model
2.2 Description of the 3D Model
2.3 OPC-UA Connection
2.4 General Configuration
2.5 Performed Tests
2.6 Experimental Results
2.7 Education Application
2.8 Future Work
3 Conclusions
References
Using Physics-Informed Machine Learning to Optimize 3D Printing Processes
1 Introduction
2 Additive Manufacturing
3 Related Work
3.1 Numerical Simulation
3.2 Machine Learning
4 Scientific Machine Learning Methodology
4.1 Deep Hidden Heat Model (DHHM)
4.2 Continuous Heat Model (CHM)
4.3 Discrete Heat Model (DHM)
5 Application to Real Datasets from SIEMENS
5.1 Sensor Data Generation
5.2 Plancks Law
5.3 Emissivity
5.4 Modeling of Heat Source
5.5 Results
6 Summary and Discussion
References
Industry 4.0 Machine-to-Machine Communication Protocols and Architectures on the Shop Floor
1 Introduction
2 Industrial Communication Systems
3 Related Work
4 Future of Industrial Networks
5 Conclusion
References
Topological Design of 3D Biopolymer Scaffolds and Their Mechanical Features
1 Introduction
2 Materials and Methods
3 Discussion of Mechanical Properties
4 Conclusion
References
Numerical Simulation of Mould-Open Microcellular Injection Moulding
1 Introduction
2 Modelling and Simulation
2.1 Part Dimension
2.2 Material
2.3 Simulation Software
2.4 Mesh Settings
2.5 Process Conditions
3 Results and Discussions
4 Conclusions
References
Design of HVAC Systems Based in Horizontal Confluent Jets Equipped in an Experimental Chamber
1 Introduction
2 Materials and Methods
2.1 Numerical Models
2.2 Horizontal Confluent Jets
3 Results and Discussion
4 Conclusions
References
Materials
The Mechanical Performance of Additive Manufactured Silica Lattice Structures
1 Introduction
2 Material and Methods
2.1 Design, 3D Printing and Post-processing of Ceramic Lattice Scaffolds
2.2 Material Characterization and Mechanical Testing
3 Results and Discussion
3.1 3D Printing and Material Characterization Process
3.2 Mechanical Behavior of Lattice Scaffolds
4 Conclusions
References
Earth as a Construction Material for Sustainable 3D Printing: Rheological Aspect
1 Introduction
1.1 Earth-Based Materials for 3D Printing
1.2 Clay Minerals: Influence on Rheology
1.3 Influence of Soil pH on the Stability and Rheology of Clay Minerals
1.4 Importance of Fibers in Enhancing the Rheology of Earth-Based Material
1.5 Estimation of the Rheology of Earth-Based Material: Squeeze Flow Rheometry
2 Conclusion
References
Mechanical and Thermal Characterization of Metal Reinforced Composites
1 Introduction
2 Experimental Procedure
2.1 Materials and Characterization Methods
2.2 Moulds and Vacuum Casting Process of the Epoxy Resin + Hardener
2.3 Mechanical Tests
2.4 Thermal Conductivity Assessment
3 Results and Discussion
3.1 Physical Properties: Density
3.2 Physical Properties: Thermal Conductivity
3.3 Mechanical Behaviour
4 Conclusions
References
An Overview of Binder Materials’ Sustainability for 3D Printing in Construction
1 Introduction
1.1 Sustainability of 3D Printing Concrete
2 Binder Materials for 3D Printing in Construction
2.1 Cementitious Matrix
2.2 Geopolymer Matrix
2.3 Earth-Based Matrix
2.4 Other Materials
3 Conclusions and Future Work
References
Mechanical Characterization of Polyamide Reinforced with Short Carbon Fibres Manufactured via FFF
1 Introduction
2 Experimental Procedure
2.1 Materials Selection
2.2 Test Specimens’ Preparation
3 Results and Discussion
3.1 Tensile and Flexural Testing
3.2 Dynamical Mechanical Analysis
4 Conclusions and Future Works
References
Applications
Aquasoft 4.0 - Administration Shell and Cloud Connection of Aquasoft
1 Introduction
2 Aquasoft Kit
3 Administration Shell
3.1 Creation of AAS and OPC-UA Server
3.2 OPC-UA Client
4 Cloud Connection
4.1 Connection of Aquasoft to Internet
4.2 Cloud Interface
5 Conclusion
References
ICM 4.0 – Injection Moulding Machine Control and Monitoring
1 Introduction
2 Case Study
2.1 IMM Control and Monitoring
2.2 IMM Digital Twin
3 Conclusions
References
Process Optimization for the Manufacturing of Individualized Ankle Foot Orthoses via Digitalization and AM
1 Introduction
2 State of the Art
2.1 Human-Parameter-Analysis
2.2 Types of Ankle Foot Orthoses
2.3 Manufacturing Process
3 Process Chain
3.1 Planning
3.2 Concept
3.3 Neuronal Network
4 Results and Conclusion
References
3D Printed Smart Luminous Artifacts
1 Introduction
2 Methodology and Approach
2.1 Design and Manufacturing of a Hybrid PCB by Multi-functional Ink-Jet Circuit Printing
2.2 Design and Manufacturing of the Electronics Housing –by Fused Filament Fabrication (FFF)
2.3 Prototyping and Testing
3 Results and Discussion
4 Conclusions
References
Smart Shoes – Current Developments and the Future Trends
1 Introduction
2 Industry 4.0 Enabling Technologies on the Development of Smart Shoes
3 Enhancing Consumers Comfort with Smart Shoes
4 Monitoring, Evaluating and Disseminating Health Problems by Smart Shoes
5 Case Studies that Promise a Great Future for Smart Shoes
6 Conclusions
References
Sustainable Water Package: Technical Characteristics and Challenges for Designers
1 Product, Package and Label
1.1 Product’s Packaging Design
1.2 Challenge for Product Design
1.3 Water Package Sustainability
2 Methodology and Methods
3 Discussion and Results
4 Conclusions
References
3D Printing and Direct Polymer Casting of Microgroove Nerve Guidance Conduits
1 Introduction
2 Materials and Methods
2.1 Design and Printing of Moulds
2.2 Casting of Microgroove Thin Films
2.3 Mould and Film Morphological Characterisation
2.4 Mechanical Characterisation
2.5 Surface Wettability Evaluation
2.6 Degradation Characterisation
2.7 Statistical Analysis
3 Result and Discussion
3.1 Microgroove Morphology
3.2 Mechanical Characterisation
3.3 Surface Wettability
3.4 Degradation Analysis
4 Conclusions
References
Investigating the Degradation Properties of Poly(ε-caprolactone) and Polyethylene Terephthalate Glycol as Biomaterials
1 Introduction
2 Material and Methods
2.1 Scaffolds Fabrication
2.2 Surface Wettability Evaluation
2.3 Accelerated Degradation Test
2.4 Morphological Evaluation
2.5 Mechanical Evaluation
2.6 Data Analysis
3 Results and Discussion
3.1 Surface Wettability Evaluation
3.2 Morphological Evaluation
3.3 Degradation Rate and Mechanical Evaluation
4 Conclusions and Future Perspectives
References
Behaviour of 3D Printed PLA Dies for Rubber Pad Forming
1 Introduction
2 Additive Manufacturing of a Die for Rubber Pad Forming Press
3 Die Behaviour during Stamping
4 Cost Analysis
4.1 Research Design
4.2 Material and Waste Cost
4.3 Economic Impact
5 Lead Time
6 Conclusions
References
Morphological Investigation of Electrospun PVDF (HFP)-Carbon Black Nanocomposites
1 Introduction
2 Methodology
2.1 Materials
2.2 Solution Preparation
2.3 Electrospinning
2.4 Morphological Characterization
2.5 Capacitance Characterization
3 Results and Discussion
4 Conclusion
References
Stainless-Steel Wire-Arc Additive Manufacturing Characterization of Single Weld Bead Deposition
1 Introduction
2 Experimental Setup: Material and Methods
2.1 Stainless-Steel ER2209
2.2 General Procedure
2.3 Samples Construction
3 Results
3.1 Method of Analysis
3.2 Impact of the Synergic Laws
3.3 Travel Speed Experiment
3.4 Wire Feeding Speed Experiment
4 Conclusions
References
Investigating Raw Earth Construction in Morocco: Actual and Future Prospects
1 Introduction
2 Traditional Earth-Based Building Techniques in Morocco
2.1 Traditional Earth-Based Building Techniques in Morocco
2.2 Cob Constructions
2.3 Earth Blocks
2.4 Embedding Fibers in Earth-Based Materials
3 Digital Fabrication of Earth-Based Materials
4 Conclusions
References
Electric Motorcycle Frame Design with Generative Design Feature for Polymer Additive Manufacturing – Concept and Prototype Validation
1 Introduction
1.1 Individual Mobility
1.2 Electric Scooters
1.3 Electric Scooter’s Frame
2 Material and Method
2.1 Steel Frame Base Model
2.2 Load Cases Considered in the Study
2.3 Material Selection
2.4 Generative Concept
3 Results
3.1 Generative Process and Geometries
3.2 Final Concept Model of Frame
4 Prototyping and Testing
4.1 Prototyping
4.2 Experimental Characterization of Frame Setup
4.3 Experimental Characterization of Frame Analysis
5 Conclusion
References
Author Index

Citation preview

Springer Tracts in Additive Manufacturing

Joel Oliveira Correia Vasco Henrique de Amorim Almeida et al. Editors

Progress in Digital and Physical Manufacturing Proceedings of ProDPM’21

Springer Tracts in Additive Manufacturing Series Editor Henrique de Amorim Almeida, Polytechnic Institute of Leiria, Leiria, Portugal

Editorial Board Members Abdulsalam Abdulaziz Al-Tamimi, Riyadh, Saudi Arabia Alain Bernard, Ecole Centrale de Nantes, IRCCyN UMR CNRS 6597, Nantes Cedex 03, France Andrew Boydston, University of Washington, Seattle, USA Bahattin Koc, Maltepe, Sabanci University, Istanbul, Türkiye Brent Stucker, Louisville, KY, USA David W. Rosen, Atlanta, GA, USA Deon de Beer, Bloemfontein, South Africa Eujin Pei , Coll of Engg, Design & Physical Sci, Brunel University London, London, UK Ian Gibson, University of Twente, Enschede, Overijssel, The Netherlands Igor Drstvensek, Faculty of Mechanical Engineering, University of Maribo, Maribor, Slovenia Joaquim de Ciurana, Girona, Spain Jorge Vicente Lopes da Silva, Ctr Info Tech Renato Archer, Campinas, São Paulo, Brazil Paulo Jorge da Silva Bártolo, Singapore, Singapore Richard Bibb, Leicestershire, UK Rodrigo Alvarenga Rezende, Araraquara, Brazil Ryan Wicker, Central Receiving, Univ of Texas at El Paso, El Paso, TX, USA

The book series aims to recognise the innovative nature of additive manufacturing and all its related processes and materials and applications to present current and future developments. The book series will cover a wide scope, comprising new technologies, processes, methods, materials, hardware and software systems, and applications within the field of additive manufacturing and related topics ranging from data processing (design tools, data formats, numerical simulations), materials and multi-materials, new processes or combination of processes, new testing methods for AM parts, process monitoring, standardization, combination of digital and physical fabrication technologies and direct digital fabrication.

Joel Oliveira Correia Vasco · Henrique de Amorim Almeida · Anabela Gonçalves Rodrigues Marto · Carlos Alexandre Bento Capela · Flávio Gabriel da Silva Craveiro · Helena Maria Coelho da Rocha Terreiro Galha Bártolo · Luis Manuel de Jesus Coelho · Mário António Simões Correia · Milena Maria Nogueira Vieira · Rui Miguel Barreiros Ruben Editors

Progress in Digital and Physical Manufacturing Proceedings of ProDPM’21

Editors Joel Oliveira Correia Vasco Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Henrique de Amorim Almeida Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Anabela Gonçalves Rodrigues Marto Computer Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Carlos Alexandre Bento Capela Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Flávio Gabriel da Silva Craveiro Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Helena Maria Coelho da Rocha Terreiro Galha Bártolo Civil Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Luis Manuel de Jesus Coelho Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Mário António Simões Correia Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Milena Maria Nogueira Vieira Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

Rui Miguel Barreiros Ruben Mechanical Engineering Department School of Technology and Management, Polytechnic Institute of Leiria Leiria, Portugal

ISSN 2730-9576 ISSN 2730-9584 (electronic) Springer Tracts in Additive Manufacturing ISBN 978-3-031-33889-2 ISBN 978-3-031-33890-8 (eBook) https://doi.org/10.1007/978-3-031-33890-8 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

The “Progress in Digital and Physical Manufacturing” book contains keynotes and papers presented at the second International Conference on Progress in Digital and Physical Manufacturing (ProDPM’21), organized by the School of Technology and Management (ESTG) of the Polytechnic Institute of Leiria (IPLeiria), from the 27th to the 29th of October 2021. This international conference aims to provide a major international forum for the scientific exchange of multidisciplinary and inter-organizational aspects performed by academics, researchers and industrial partners in order to exchange ideas in the field of digital and physical manufacturing and related areas. It represents a significant contribution to the current advances in industrial digital and physical manufacturing issues as it contains topical research in this field. The ProDPM’21 conference expects to foster networking and collaboration among participants to advance the knowledge and identify major trends in the field. The conference addresses industrial challenges focused on current market demands and actual technological trends, such as mass customization, new business and industrial models or predictive engineering. Its contribution in science and technology developments leads to more suitable, effective and efficient products, materials and processes, generating added value for the industry and promoting the awareness of the role and importance of the digital and physical manufacturing development in the society. This book is, therefore, an essential reading for all of those working on digital and physical manufacturing, promoting better links between the academia and the industry. The conference papers will cover a wide range of important topics like additive manufacturing, biomanufacturing, advanced and smart manufacturing technologies, rapid tooling, micro-fabrication, virtual environments, simulation and 3D CAD and data acquisition, materials and collaborative design. We are deeply grateful to the keynote speakers, authors, participants, reviewers, the International Scientific Committee, Session chairs, student helpers and administrative assistants, for contributing to the success of this conference. Joel Vasco Henrique Almeida Conference Co-chairs

Conference Committees

Conference Co-chairs Joel Oliveira Correia Vasco

Henrique de Amorim Almeida

Mechanical Engineering Department, School of Technology and Management, Polytechnic Institute of Leiria, Leiria, Portugal [email protected] Mechanical Engineering Department, School of Technology and Management, Polytechnic Institute of Leiria, Leiria, Portugal [email protected]

Organizing Committee Alexandrino José Marques Gonçalves Anabela Gonçalves Rodrigues Marto Carlos Alexandre Bento Capela Eliseu Manuel Artilheiro Ribeiro Flávio Gabriel da Silva Craveiro Helena Maria Coelho da Rocha Terreiro Galha Bártolo Liliana Coutinho Vitorino Luís Manuel de Jesus Coelho Luís Miguel Ramos Perdigoto Maria Leopoldina Mendes Ribeiro de Sousa Alves Mário António Simões Correia Milena Maria Nogueira Vieira Rui Miguel Barreiros Ruben

Scientific Committee Abdulsalam Al-Tamimi Alain Bernard António Augusto Magalhães Cunha António Manuel Ramos António Pouzada Bernhard Mueller

King Saud University, Saudi Arabia Centrale Nantes, France University of Minho, Portugal University of Aveiro, Portugal University of Minho, Portugal Fraunhofer Institute for Machine Tools and Forming Technology IWU, Germany

viii

Conference Committees

Bopaya Bidanda Brent Stucker Dachamir Hotza David Rosen De Beer Deon Eujin Pei Fernando A. Lasagni Giorgia Franchin Ian Campbell Igor Drstvensek Jan T. Sehrt Jean-Yves Hascoet Jenny Chen João Manuel R. S. Tavares Joaquim de Ciurana Jorge Vicente Lopes da Silva José Simões Júlio César Viana Khalid Zarbane Luigi Galantucci Martin Schaefer Maryam Hojati Michael Friedrich Zäh Ming C. Leu Nickolas Sapidis Olivier Jay Paulo Bártolo Paulo Fernandes Renato Manuel Natal Jorge Rodrigo Alvarenga Rezende Steinar Killi Swee Leong Sing Tahar Laoui Tatjana Spahiu Terry Wohlers Tugrul Ozel

University of Pittsburgh, USA ANSYS Inc, USA Federal University of Santa Catarina, Brazil Georgia Institute of Technology, USA Central University of Technology, South Africa Brunel University, UK Advanced Center for Aerospace Technologies, Spain University of Padua, Italy Loughborough University, UK University of Maribor, Slovenia Ruhr University Bochum, Germany Ecole Centrale de Nante, France 3DHEALS, USA University of Porto, Portugal University of Girona, Spain CENPRA, Brazil Escola Superior de Artes e Design, Portugal University of Minho, Portugal University of Casablanca, Morocco Politecnico di Bari, Italy Siemens AG, Germany University of New Mexico, USA acatech – National Academy of Science and Engineering, Germany Missouri University of Science and Technology, USA University of Western Macedonia, Greece Faddtory, Belgium University of Manchester, UK Instituto Superior Técnico, Portugal University of Porto, Portugal University of Araraquara, Brazil Oslo School of Architecture and Design, Norway Nanyang Technological University, Singapore King Fahd University of Petroleum and Minerals, Saudi Arabia Polytechnic University of Tirana, Albania Wohlers Associates, USA Rutgers University, USA

Conference Committees

ix

Acknowledgments and Sponsors The conference co-chairs and organizing committee wish to acknowledge the support and sponsorship given to the organization of the ProDPM’21—International Conference on Progress in Digital and Physical Manufacturing:

Official Sponsors

Event Sponsors

Event Promoters

Institutional Sponsors

Invited Lectures

The conference had the privilege of including in the scientific programme the following world-renowned speakers:

Alain Bernard École Centrale de Nantes, France Prof. Alain Bernard, 60, graduated in 82, PhD in 89, was Associate Professor, from 90 to 96 in Centrale Paris. In 96, he got Full Professor position in University Nancy I, in the “Integrated Design and Manufacturing” team, and moved to Centrale Nantes in 01 where he was Dean for Research (07– 12). He is researcher in Digital Sciences Laboratory (LS2NUMR CNRS 6004) in the “Systems Engineering” (IS3P) team. Recent research topics are KBE applied to computeraided decision-making systems for additive manufacturing. He is Vice-President of the French Additive Manufacturing Association (AFPR), CIRP Fellow and Member of the French National Academy of Technologies.

Eujin Pei Brunel University, UK Dr. Eujin Pei is the Director for Postgraduate Research and Programme Director for the BSc Product Design and BSc Product Design Engineering programmes at Brunel University London. He is a Chartered Engineer (CEng) and a Chartered Technological Product Designer (CTPD) with the Institution of Engineering Designers. He is the Convenor of the International Organization for Standardization ISO/TC261/WG4 committee that is responsible for Data Transfer and Design Standards for Additive Manufacturing and Chair for the British Standards Institute BSI/AMT/008 for Additive Manufacturing. His research interest centres on Functionally Graded Additive Manufacturing and 4D printing.

xii

Invited Lectures

David Hsu CoreTech System, Taiwan Dr. David Hsu is the Deputy CEO of CoreTech System. He received his PhD degree in Chemical Engineering from National Tsing-Hua University (Taiwan) in 1995. He is one of the founders of CoreTech System and lead the R&D team to develop Moldex3D for 25 years. He has over 20 publications and over 15 patents and pending patents in the fields of moulding processes simulation, material modelling and smart manufacturing digital twins. In 2019, he and his team received the “Aoki Katashi Innovation Award” from the Japan Society of Polymer Processing (JSPP) for their contribution in the development of microcellular foam injection moulding simulation technology.

Deon de Beer Central University of Technology South Africa Deon started his career at UCOR (Now NECSA) in 1979 and joined the Technikon Free State (TFS) in 1997, where he held several positions. During his tenure at the TFS/CUT, he spent a sabbatical at MATTEK, CSIR in 1995, to start a research project on Rapid Prototyping, to assess industry perceptions and needs by focusing on the role of Rapid Prototyping to support Concurrent Engineering in South Africa, under the leadership of the late Dr Neville Comins and Prof Willie du Preez. Involvement in Rapid Prototyping/Additive Manufacturing R&D changed his future, and led to various other opportunities at the CUT, VUT and NWU, before returning to CUT in 2018, in addition to be involved in new national initiatives and networks locally and internationally. He believes in empowering others while maintaining personal excellence and leading by example. One of his personal objectives is continuous promotion and education of AM and its innovation and commercialization in SA and beyond.

Invited Lectures

xiii

Giorgia Franchin University of Padova, Italy Dr Giorgia Franchin is a Research Fellow at the University of Padova in Italy where she is focusing on additive manufacturing of ceramic materials. Her PhD research was on innovative additive manufacturing technologies for ceramic materials to enable new properties and complex shapes. She was working on direct ink writing of preceramic polymers to produce bioceramic scaffolds and ceramic matrix composites; and of geopolymers to develop advanced components for air and water filtration purposes.

Jenny Chen 3DHEALS, USA Jenny Chen, MD, is the Founder and CEO of 3DHEALS, a company focusing on education, communication and investments in the space of bioprinting, regenerative medicine, healthcare applications using 3D printing since 2015. She is trained as a neuroradiologist, and she holds degrees in both medicine and radiology from the David Geffen School of Medicine at UCLA and completed a neuroradiology fellowship at Harvard Medical School. She served as Adjunct Clinical Faculty in neuroradiology at Stanford University Medical Center for more than seven years until recently to focus more on works with 3DHEALS. She serves as a mentor and an advisor to start-ups in the space. Her interests lie in the applications of 3D printing and bioprinting in health care, automated biology, and has a vision of a decentralized and personalized healthcare delivery system for our near future.

xiv

Invited Lectures

Marco Göbel TRUMPF, Germany Marco Goebel is a diploma physicist with more than 13 years of experience in industrial laser applications. His specialties are laser metal deposition and additive manufacturing. After graduation, he worked at the Fraunhofer Institute for Laser Technology ILT in the Laser Metal Deposition group, where he conducted and managed industrial research and development projects as a research associate. In industry, he worked for a large laser manufacturer and machine builder. In this position, he was responsible for integration solutions and provided application consultancy for various laser material processing technologies. For more than 2 years, he has been working for TRUMPF in the Industry Management department.

Omar Fergani Atotech, Germany Dr. Fergani is a heading the digital manufacturing division at Atotecha— leading technology provider for the electronic and semiconductor manufacturing. His team is researching, developing and deploying Industry 4.0 solutions based on AI and IoT. Additionally, he is leading the additive manufacturing unit focusing on implementing a decentralized manufacturing approach enabling spare parts on demand as well as a digital inventory. Before joining Atotech, he was an R&D director at Siemens in charge of digital manufacturing technologies. He is a lecturer in multiple universities and board member of multiple technology start-ups. He is a holder of a Ph.D. in mechanical engineering from Georgia Tech and NTNU. He was also awarded the outstanding young manufacturing engineer by SME.

Invited Lectures

xv

Paulo Bártolo University of Manchester, UK Paulo Bártolo is Professor of Advanced Manufacturing and Head of the Manufacturing Group at the School of Mechanical, Aerospace and Civil Engineering, University of Manchester. He is the University’s Industry 4.0 Academic Lead, team leader of the Industry 4.0 societal challenge at Digital Futures and sits on the Management Board of the EPSRC & MRC Centre for Doctoral Training in Regenerative Medicine. He is Professor at the Advanced Manufacturing Group at the Tecnologico de Monterrey, at Nanyang University, and member of CIAUD (at University of Lisbon). He is a Fellow of CIRP, advisor of the Brazilian Institute of Biofabrication and several UK and International Funding Agencies and received a commendation and pubic recognition from the Portuguese Government. He is the Founding Editor of Virtual and Physical Prototyping Journal and Editor-in-Chief of Biomanufacturing Reviews.

Steffen Sickinger TRUMPF, Germany Working for TRUMPF over 11 years in several positions. Starting as apprentice for industrial mechanic at TRUMPF and now for the last 4 years in the sales department of additive manufacturing, responsible for western Europe and American markets. Academic background is in business administration and engineering—international management for the bachelor’s degree and business consulting and digital management for master’s degree and spend 7 months in the USA for TRUMPF.

xvi

Invited Lectures

Tatjana Spahiu Kosova Polytechnic University of Tirana, Albania Dr. Tatjana Spahiu is Lecturer at the Department of Textile and Fashion, Faculty of Mechanical Engineering at Polytechnic University of Tirana. Her research is focused on different applications of 3D technologies for fashion products. During her PhD, she implemented for the first time 3D scanning technology for digital anthropometry in Albania as a pilot project. She continues to develop and create fashion products including garments, footwear, accessories through 3D technologies. Through additive manufacturing technology as a sustainable way of production she continues to use it for its products. She has more than 30 papers presented and published in various international scientific conferences and journals.

Terry Wohlers Wohlers Associates, Inc., USA Terry Wohlers is President of Wohlers Associates, Inc., an independent consulting firm he founded 32 years ago. He has authored more than 421 books, articles and technical papers on product development and manufacturing and has given 155 keynote presentations on five continents. In 2004, he received an Honorary Doctoral Degree in Mechanical Engineering from Central University of Technology in Bloemfontein, South Africa. In 2005, he became a Fellow of the Society of Manufacturing Engineers (SME). In 2016, he became an adjunct professor at RMIT University in Melbourne, Australia. For 24 years, he has been a principal author of the Wohlers Report, an annual worldwide publication focused on additive manufacturing and 3D printing.

Conference Awards

With the support of the Scientific Committee and the Conference Sponsors, the following awards were presented at the ending ceremony of the conference: The Best Paper Award was awarded to authors whose work represents groundbreaking research in their respective areas: • The authors Fabian Soffel, Sergei Egorov, Dominik Keller & Konrad Wegener with the paper intitled “CAD/CAM process chain for hybrid additive manufacturing” – Award presented by Moisés Domingues, the CEO of the company CODI.

The Best Young Researcher was awarded to young researchers, considering both the paper quality (assessed by the Scientific Committee) and the oral presentation quality (feedback given by session chairs at the conference venue): • The authors Jiong Yang, Wajira Mirihanage & Paulo Bartolo with the paper intitled “Novel Extrusion Based Co-Axial Printing Head for Tissue Engineering” – Award presented by Fabiana Guarda, the CEO of the company Lansys. • The authors Xiyao Ni, Thomas Kendall & Paulo Bartolo with the paper intitled “Modelling the Material Removal Process of Turbulent Jet Electrochemical Machining of Copper” – Award presented by Fabiana Guarda, the CEO of the company Lansys.

The Best Industrial Award was awarded to authors whose applied research represents a significant scientific/technical contribution towards the industrial community:

xviii

Conference Awards

• The authors Carel Plekker, Francois Kuys, William Allan Kinnear & Jacques Combrinck with the paper intitled “Rapid product development: A decisionmaking matrix for the manufacturing of injection mould inserts for small batch production” – Award presented by João Ferreira, the CEO of the company S3D.

Contents

Keynote and Workshop Integration of Additive Manufacturing in Production Systems . . . . . . . . . . . . . . . . Alain Bernard

3

How Virtual and Augmented Reality Can Boost Manufacturing . . . . . . . . . . . . . . Anesio Neto, Eduarda Abrantes, Carlos Rabadão, Andreia Jesus, Gustavo Reis, Filipe Gonçalves, Gabriel Evangelista, Ricardo Antunes, and Vítor Ferreira

12

Advanced Manufacturing Technologies CAD/CAM Process Chain for Hybrid Additive Manufacturing . . . . . . . . . . . . . . . Fabian Soffel, Sergei Egorov, Dominik Keller, and Konrad Wegener On the Quality of Electron Beam Melted Thin-Walled Parts with Curved Surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gabriele Piscopo, Alessandro Salmi, Eleonora Atzeni, Luca Iuliano, Giovanni Marchiandi, Adriano Nicola Pilagatti, Giuseppe Vecchi, and Mirna Poggi

41

47

Computational Origami Based Design in 4D Printing . . . . . . . . . . . . . . . . . . . . . . . Mohamed H. Hassan, Jatin Sharma, and Paulo Bartolo

54

Novel Extrusion Based Co-axial Printing Head for Tissue Engineering . . . . . . . . Jiong Yang, Wajira Mirihanage, and Paulo Bartolo

60

Localization and Control of a Mobile Robot for Additive Manufacturing . . . . . . Abdullah Alhijaily, Zekai M. Kilic, and Paulo Bartolo

72

Development of a Large Size 3D Delta Printer for Advanced Polymers . . . . . . . . D. Pereira, M. Leite, M. Ferreira, D. Machado, R. Dionísio, and R. A. Cláudio

82

Modelling the Material Removal Process of Turbulent Jet Electrochemical Machining of Copper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiyao Ni, Thomas Kendall, and Paulo Bartolo

95

Size Matters, Designing for Larger AM Products! . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Steinar Killi and Magne Solvaag Mathisen

xx

Contents

Design and Green Manufacturing - CAD and 3D Data Acquisition Technologies Rapid Product Development: A Decisionmaking Matrix for the Manufacturing of Injection Mould Inserts for Small Batch Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 A. C. Plekker, F. A. Kuys, W. A. Kinnear, and J. Combrinck A Methodical Approach to Product Development in 4D Printing Using Smart Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Stefan Junk, Henning Einloth, and Dirk Velten Product Design for the Circular Economy: A Design Process for Footwear . . . . . 138 Dirk Loyens, Shujoy Chakraborty, and Diogo Pimenta Design of Playful-Pedagogical Objects for Learning and Development of Preschool-Aged Children with Autism Spectrum Disorders (ASD) . . . . . . . . . 152 Inês Mimoso, Luciana Barbosa, and António Marques 3D Pine Tree Geometry Design in Forest Fire Environments . . . . . . . . . . . . . . . . . 163 Eusébio Conceição, João Gomes, M. Manuela Lúcio, Jorge Raposo, Domingos XavierViegas, and M. Teresa Viegas Recycled Reinforced PLA as Ecodesign Solution for Customized Prostheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Marcelo Gaspar, Miguel Ferraz, Armando Ramalho, Joel Vasco, and Carlos Capela Digital Manufacturing and Simulation Systems Virtualization and Optimization of Processes in Industry 4.0 . . . . . . . . . . . . . . . . . 197 Jérôme Rodrigues and Eliseu Ribeiro Using Physics-Informed Machine Learning to Optimize 3D Printing Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 Benjamin Uhrich, Martin Schäfer, Oliver Theile, and Erhard Rahm Industry 4.0 Machine-to-Machine Communication Protocols and Architectures on the Shop Floor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 Marcella Cavalcanti, Hugo Costelha, and Carlos Neves Topological Design of 3D Biopolymer Scaffolds and Their Mechanical Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Igor Shishkovsky, Oleg Dubinin, and Stepan Konev

Contents

xxi

Numerical Simulation of Mould-Open Microcellular Injection Moulding . . . . . . 242 Yifei Ding and Paulo Bartolo Design of HVAC Systems Based in Horizontal Confluent Jets Equipped in an Experimental Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Eusébio Conceição, Vasco Correia, Mª Inês Conceição, Mª Manuela Lúcio, João Gomes, and Hazim Awbi Materials The Mechanical Performance of Additive Manufactured Silica Lattice Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 N. Kladovasilakis, T. Kontodina, K. Tsongas, E. M. Pechlivani, D. Tzetzis, and D. Tzovaras Earth as a Construction Material for Sustainable 3D Printing: Rheological Aspect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 Arpan Joshi, Philippe Poullain, Flávio Craveiro, and Helena Bártolo Mechanical and Thermal Characterization of Metal Reinforced Composites . . . . 281 Lécio Lourenço, Flávio Craveiro, Joel C. Vasco, and Carlos Capela An Overview of Binder Materials’ Sustainability for 3D Printing in Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 Ye¸sim Tarhan, Flavio Craveiro, and Helena Bartolo Mechanical Characterization of Polyamide Reinforced with Short Carbon Fibres Manufactured via FFF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 João Brites, Flávio Craveiro, Joel C. Vasco, and Carlos Capela Applications Aquasoft 4.0 - Administration Shell and Cloud Connection of Aquasoft . . . . . . . 315 Rodrigo Marques and Eliseu Ribeiro ICM 4.0 – Injection Moulding Machine Control and Monitoring . . . . . . . . . . . . . 322 Guilherme Santos, Miguel Lopes, Jérôme Rodrigues, Eliseu Ribeiro, and Joel C. Vasco Process Optimization for the Manufacturing of Individualized Ankle Foot Orthoses via Digitalization and AM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330 Lydia Mika, Arthur Hilbig, Ulrike Gebhardt, Fatemeh Mehdipour, Paul Naake, Stefan Holtzhausen, and Ralph Stelzer

xxii

Contents

3D Printed Smart Luminous Artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 V. Papadopoulou, T. Kontodina, E. M. Pechlivani, G. Kastrinaki, A. Asimakopoulou, I. Tzitzios, D. Ioannidis, and D. Tzovaras Smart Shoes – Current Developments and the Future Trends . . . . . . . . . . . . . . . . . 346 Tatjana Spahiu, Henrique Almeida, Panagiotis Kyratsis, Antonio Jimeno-Morenilla, and Sarghie Bogdan Sustainable Water Package: Technical Characteristics and Challenges for Designers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 Raquel Antunes, Henrique Almeida, Liliana Vitorino, and Fernanda Carvalho 3D Printing and Direct Polymer Casting of Microgroove Nerve Guidance Conduits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 Hexin Yue, Cian Vyas, and Paulo Bartolo Investigating the Degradation Properties of Poly(ε-caprolactone) and Polyethylene Terephthalate Glycol as Biomaterials . . . . . . . . . . . . . . . . . . . . . 379 Yanhao Hou, Weiguang Wang, and Paulo Bartolo Behaviour of 3D Printed PLA Dies for Rubber Pad Forming . . . . . . . . . . . . . . . . . 388 R. A. Cláudio, P. Cardoso, H. Ferreira, V. Alcácer, and J. Simões Morphological Investigation of Electrospun PVDF (HFP)-Carbon Black Nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 Abdalla M. Omar, Cian Vyas, Mohamed H. Hassan, and Paulo Bartolo Stainless-Steel Wire-Arc Additive Manufacturing Characterization of Single Weld Bead Deposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 Ricardo Viola, Mário S. Correia, Leopoldina Alves, Pierre Michaud, and Anaïs Domergue Investigating Raw Earth Construction in Morocco: Actual and Future Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419 Karim Fahfouhi, Henrique Almeida, Dino Freitas, Flávio Craveiro, and Helena Bártolo Electric Motorcycle Frame Design with Generative Design Feature for Polymer Additive Manufacturing – Concept and Prototype Validation . . . . . . 430 Pedro Domingues, Carlos Relvas, and António Ramos Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

Keynote and Workshop

Integration of Additive Manufacturing in Production Systems Alain Bernard(B) Ecole Centrale de Nantes, LS2N UMR CNRS 6004, Nantes, France [email protected]

1 Introduction Historically, Additive Manufacturing is not considered as a production technology even if many examples are available in different application fields [1]. More than 20 years ago, some specific and niche sectors have adopted additive manufacturing technologies as part of a complete value chain, from 3D scanning and/or CAD modeling to post-processing. With the progress of reliability of additive manufacturing technologies and with respect to the increase of expertise of companies in this field during the last 20 years, some alternative visions have been drawn concerning new designs and new ways of producing products, by integrating functions, by optimizing material quantity, by considering a coherent association of several technologies with respect to a given set of requirements (and not only additive manufacturing technologies). In addition, AM became cost-effective [2] and especially for spare parts. For such parts, one way of progress is to produce on demand parts when and where you need then, with up-to-date digital models. All this progress is supported by national and international associations, like France Additive. In this short paper, France Additive is first briefly introduced. Then, Additive Manufacturing is positioned as a key player of production systems in the context of the factory of the future and Industry 4.0. Some particular key issues are then introduced with several concrete illustrations. In the last section, examples of industrial facilities and production structures are introduced and commented. Then a conclusion section allows summarizing the content of the paper and highlighting some expected progress for the future.

2 A Key Actor: France Additive France Additive is a non-profit association, dedicated to Additive Manufacturing and created in 1992 with another name, French Association for Rapid Prototyping (AFPR). In a recent evolution, many activity groups have been created and devoted to the different topics supported by the 180 + members. The main goal of France Additive is to federate research and technical centers, education actors, industrial companies, service providers and final users. The scope of topics of interest is 360° in an open community, from R&D, applications, technologies (physical and digital), materials, skills and education, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 3–11, 2023. https://doi.org/10.1007/978-3-031-33890-8_1

4

A. Bernard

finance, supply chain, economic growth. Recently, the French government decided to create a new branch strategic contract named “Solutions Industrie du Futur”. France Additive is in charge of coordinating the effort in the field of Additive Manufacturing. Part of the action plan is to accelerate the development of AM in France, to increase the competitiveness of French AM actors activities, to improve the cooperation between all the actors of the ecosystem within complete value chains, to boost the technical investments in the field of AM.

3 A Key Player: Additive Manufacturing At a worldwide level, Additive Manufacturing activities have increased significantly during the last 15 years, mostly. In this context, many technologies have been developed, many standards have been proposed and many companies have adopted AM. But adopting AM is not only buying and using a AM machine. This is a complete transformation of the value chain, from design to post-treatments, with a real systemic vision. Managing product lifecycle is mandatory, including all the dimensions of the lifecycle (mainly product, process, organization), taking into account the material lifecycle and trying to optimize the material consumption and the material recycling when possible. Two parallel flows have to be “synchronized”: physical and digital ones. As shown on the following Fig. 1, many issues have to be taken into account in order to achieve the requested and expected products at the end of the complete value chain process.

Fig. 1. Global schema of a systemic vision of AM [3].

Behind this schema, there are many actions and many actors that have to be considered, those who are on the critical path dedicated to the direct production of the product, and many other activities that could be considered as services all along the complete

Integration of Additive Manufacturing in Production Systems

5

development process of a given product. The following Fig. 2 summarizes the set of the main activities and actors concerned for metal layer-based fusion manufacturing value chain.

Fig. 2. Interoperability between actors along the value chain.

4 Key Issues of Additive Manufacturing in Relation with Industrial Production In this context, one major question is to decide to adopt and to integrate AM as a production solution in a company for regular or specific productions. The main issue is to consider all the dimensions that concern the new way of designing, of producing, of managing products along their lifecycle. In this paper, only five key factors will be considered and illustrated. The following Fig. 3 summarizes those five factors: innovative creation of innovative products, spare parts, digital logistics and distributed production, direct production, production on-demand and zero stock. Each of them will be briefly commented and illustrated with some particular examples. 4.1 Innovative Creation of Innovative Products Design or redesign of parts in the context of a value chain based on AM has new great benefits in the field of design methodology. Design for Additive Manufacturing [4, 5] allows optimizing different factors, in particular part weight by putting the just enough material at the right place with for example topological optimization. But one major advantage is the integration of functions and the use of multi-material for the same part during a given production. In the following Fig. 4, three examples are proposed, two directly relate to function integration, reducing dramatically the number of parts in a

6

A. Bernard

Fig. 3. Five key factors to be considered for AM integration and success.

unique one and considering new opportunities to design and manufacture tubes, channels or hydraulic blocs. Instead of drilling blocs of materials, the idea is to put material around the tubes and channels and to connect them with free forms very easy to fabricate with AM. Using multi-material is also an opportunity to get in a unique part different material characteristics instead of assembling different parts, each of them dedicated to a given function and fabricated with a different material. In that case, design tools have to propose additional functions because usually, material is a unique characteristic of a given part.

Fig. 4. New designs with integration of functions and use of multi-material technologies.

And as it may be seen I the following Fig. 5, design may be as complex as needed with respect to compactness of shapes of parts. New opportunities are offered to heat exchangers ad also to hydraulic blocs, as previously mentioned. At the same time, what is expected is to get a design without any support because some shapes are just inaccessible to be cleaned. Applied mathematics and optimization software modules have been recently developed/improved and proposed to designers in order to really get all the benefit of using AM in a given value chain. Obviously, topological optimization is seen as one major tool to define or predefine the 3D complex shapes of the parts. What is still complex is to take into account all influential factors from the manufacturing

Integration of Additive Manufacturing in Production Systems

7

technologies when designing the part and not only the constraints related to the future use of the product. Major progresses in this field are expected during the next ten years based on a better understanding of the physical phenomena and on the proposition on new optimization models and approaches.

Fig. 5. Topological optimization of shapes, optimized heat exchanges and hydraulic blocs.

In addition, all these new opportunities of design and manufacture allow defining aesthetic and artistic designs, unique or with a set of customized / personalized parts, all different, without any additional cost because of complexity of diversity, even if produced during the same production batch. As shown in the following Fig. 6, some interesting applications have been industrialized during the last twenty years, like hearing aids for example.

Fig. 6. Examples illustrating zero additional cost for complexity and diversity.

8

A. Bernard

4.2 4–2 Spare Parts Manufacturing spare parts is a very promising opportunity for AM application. This can be applied because it is sometimes impossible and always costly to maintain an inventory of spare parts along many years. The main idea is to produce spare parts when and where needed, transforming physical flow into digital flow. The following Fig. 7 shows some examples in different fields. A recent evolution would be that the CAD files of some critical parts of appliances, like washing machines, could be accessible for free in order to let customer manufacture them by themselves. But the main industrial challenge concerns spare parts for many industrial fields like transports, energy, aeronautics and space, agriculture, etc. In fact there is a double challenge: to be able to repair parts or if not possible to manufacture new ones.

Fig. 7. Examples of spare parts in different fields.

4.3 4–3 Digital Logistics and Distributed Production As mentioned before, manufacturing of spare parts is expected when and where it is needed. But at a more general level, the same concept could be developed for many productions. When it is possible, in order to avoid to transport parts/products and to minimize carbon foot print, production could be processed in a distributed manner and not centralized. Some software platforms and market places are able to propose a complete integration of data from design to end-of-life. At the same time, it is necessary to secure the digital chain in order to be sure that the manufacturing process of a give part/product is the right one. This is crucial for parts that are certified in given field, and it is also highly expected when considering counterfeiting trademarked goods. In that case, block chain technology may help in improving the complete traceability and certification of a given specific process of a part/product all along the supply chain. An example of solution available on the market is given in Fig. 8.

Integration of Additive Manufacturing in Production Systems

9

Fig. 8. Example of a block chain solution (MainChain from Vistory company).

4.4 4–4 Direct Production As mentioned before, many operational solutions are available to allow industry to distribute production around the World, based on digital logistics and block chain solutions for example to secure the manufacturing processes, instead of centralizing production and distributing the physical products. So, what is really expected for the future is direct production based on AM technologies and value-chains. Many examples of parts in different fields illustrate the direct production capabilities offered by AM, not only for parts but also for tools and toolings. One of the main issues related to productivity. Metallic parts could be obtained using AM but manufacturing time is still very long. Some recent solutions have been proposed by machine manufacturers to decrease manufacturing time, to increase the number of energy sources (for example 4 to 12 lasers in laser-bed fusion machine for metals), to optimize the management of energy distribution (de-focalization, beam shaping for example) or more generally, to minimize the building time for each layer. Decreasing manufacturing time while decreasing also material cost while improving the design to minimize cleaning post-process operations, all these progresses let hope a bright future for industrial applications of AM in a near future (5 to 10 years) in many fields. But this is not so simple. Design methods and tools have to be adapted and designers have to be trained for new skills and new design principles. New specific skills are also needed in the shop floors for manufacturing, control and maintenance. 4.5 4–5 Production on Demand – Zero Stock The general tendency is to use appropriate manufacturing structures able to be efficient, resilient, flexible, agile and adaptable. But when these characteristics are mentioned, it is not so easy to convert them into reality. This means that production on demand should be mostly adaptable to personalization issue. Production systems which include AM machines need to develop new automation functions in order to increase productivity

10

A. Bernard

even if what has to be produced is not repeatable. However, in many cases, the products are similar, even always the same with some few customized items. So, this is difficult, often impossible, to plan production batches very early. Efficient design and production management systems are the keys of a great success for such kind of production. Then, the post-treatment and shipping systems have to be also efficient in order to minimize the global lead time and to increase customer satisfaction.

5 Conclusions Many other factors are I favor of the integration of Additive Manufacturing in production systems. However, a lot of progress is expected in a near future to really help in switching to a real economic reality of AM usage. But R&D centers should not stop their efforts in favor of continuous improvements, in particular concerning productivity and robustness of the global value chain processes. As mentioned in this paper by Schmidt et al. [6], enhancement of material portfolio is a major expectation in metallic parts production, including graded materials and multi-material alloys. Predefinition of global processes, including hybrid combinations of conventional and non-conventional technologies, is also a challenge with respect to the best compromise between all major requirements, in particular lead time, cost and quality. New key performance indicators should be taken into account by optimization and simulation modules in order to really anticipate the best strategies in a given context. Learning about physical phenomena but, more widely, about real limits of technologies is absolutely necessary at a general level and in specific contexts and application fields. Specific in-process control systems could help for closed-loop auto-correction systems, helping in adjusting the process parameters in real time, in detecting and correcting main well-known possible defaults along manufacturing process. This could provide real quality assurance systems helping for certification matters. When speaking of production, this also means large volumes of production in many cases. So, enlarging production capacities as well as accelerating material transformation are two main goals that are proposed for future developments. Prost-process and finishing technologies are essential to assure the final quality of the product. Many progresses have to be done in this field in order to minimize the cost and efforts dedicated to these critical phases of the global process. And finally, control technologies, mostly non-destructive ones, have to be improved as well as quality assurance approaches. So, definitively, AM is a real key player for future production systems. Real changes will allow creating products only manufacturable with AM technologies. It is already the case today but, as soon as much more actors will be confident in the robustness and productivity of AM technologies, associated to efficient post-processes, a significant evolution will be possible at a large scale. As described in this short paper, many advantages are offered by AM technologies and AM-based value chains. But this evolution will come day after day, in parallel of other technologies that have already shown their efficiency for production. Construction, medical sector, and other sectors have already started to adopt AM but significant progresses have also to be seen with the specific regulations and certifications. Obviously, Additive Manufacturing is a field of interest that is promised to a bright future.

Integration of Additive Manufacturing in Production Systems

11

References 1. Ahuja, B., Karg, M., Schmidt, M.: Additive manufacturing in production – challenges and opportunities, laser 3D manufacturing II. In: Helvajian, H., Piqué, A., Wegener, M., Gu, B. (eds.) Proceedings of SPIE, vol. 9353, 935304 (2015). SPIE · CCC code: 0277-786X/15/$18. https://doi.org/10.1117/12.208252 2. Thomas, S., Gilbert, S.W.: Cost and cost-effectiveness of additive manufacturing: a literature review and discussion. NIST special publication 1176. https://doi.org/10.6028/NIST.SP.1176 3. Martinez, L.: A generic systemic vision and modelling of additive manufacturing processes. Master thesis, Ecole Centrale de Nantes, IS3P team of LS2N UMR CNRS 6004 (2016) 4. Thompson, M.K., et al.: Design for additive manufacturing: trends, opportunities, considerations, and constraints. CIRP Ann. 65(2), 737–760. https://doi.org/10.1016/j.cirp.2016. 05.004 5. Vaneker, T., Bernard, A., Moroni, G., Gibson, I., Zhang, Y.: Design for additive manufacturing: framework and methodology. CIRP Ann. 69(2), 578–599. https://doi.org/10.1016/j.cirp.2020. 05.006 6. Schmidt, M., et al.: Laser-based additive manufacturing in industry and academia. CIRP Ann. 66(2), 561–583. https://doi.org/10.1016/j.cirp.2017.05.011

How Virtual and Augmented Reality Can Boost Manufacturing Anesio Neto1 , Eduarda Abrantes2(B) , Carlos Rabadão2 , Andreia Jesus3 , Gustavo Reis2 , Filipe Gonçalves3 , Gabriel Evangelista3 , Ricardo Antunes3 , and Vítor Ferreira3 1 VRARA, Virtual Reality and Augmented Reality Association Portugal, Lisbon, Portugal 2 CIIC, ESTG, Polytechnic of Leiria, Morro do Lena, Leiria, Portugal

[email protected] 3 ESTG, Polytechnic of Leiria, Morro do Lena, Leiria, Portugal

Abstract. Immersive technologies, namely Augmented Reality and Virtual Reality, have immense capacity to improve manufacturing processes in any kind of industry. These are not newly created technologies, some documented experiments date back to the 1960’s, initially with military applications in the United States of America and also development in Universities. Recently an increase in adoption, due to considerable reduction not only in the cost of production of the electronic and optical equipment, but also in the cost and ease of developing experiences with software for use, often free of charge. This article presents not only the theory, existing equipment and use cases of each of these two immersive technologies, but also, through five different themes in the industries (Simulation and Training; Remote Field Services; Team Collaboration; Sales and Marketing; Product or Project Design), the article presents possible use cases and an indication of how to plan and implement experiments with effective results. Keywords: Augmented Reality · Virtual Reality · Industry 4.0 · Manufacturing · Training · Field Service · Remote Collaboration · Information Technology

1 Introduction Virtual Reality (VR) and Augmented Reality (AR) provide the common human eye with the power of a superhero. That’s what science fiction got us used to, with great prominence in the 80’s, and which is now increasingly present in our daily lives. Technological evolution at this level is currently growing at a very high rate, providing an investment by many companies producing these equipment, also content production companies and companies seeking this revolution to apply in their daily routines, to their products or even provide an enriched experience to customers. Thus, and ten years after the desire for digital transformation in the industry, the well-known “industry 4.0”, we see its gradual implementation, growing at the pace of technological growth in equipment and communications. Studies prove its effectiveness with results that greatly enhance the success of companies at different levels. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 12–37, 2023. https://doi.org/10.1007/978-3-031-33890-8_2

How Virtual and Augmented Reality Can Boost Manufacturing

13

Also known as immersive technologies, AR and VR have immense capacity to improve manufacturing processes in any kind of industry. These are not newly created technologies, some documented experiments date back to the 1960’s, initially with military applications in the United States of America and also development in Universities and in Academy. Recently an increase in adoption, due to considerable reduction not only in the cost of production of the electronic and optical equipment, but also in the cost and ease of developing experiences with software for use, often free of charge. Taking into account the current scenario and regarding the need to involve these technologies in industrial manufacturing, this workshop was proposed in order to spread knowledge about the technologies of Virtual Reality and Augmented Reality available in the market for the industrial manufacturing, presenting their advantages and disadvantages, showing the specificities and applicability of each one; well-known and successful case studies that make use of immersive technologies were also presented, and finally their industrial applicability showing existing cases and how these technologies can be implemented in manufacturing in order to be successful. This article presents not only the theory, existing equipment and use cases of each of VR and AR, but also, through five different themes in the industries (Simulation and Training; Remote Field Services; Team Collaboration; Sales and Marketing; Product or Project Design), it also presents possible use cases and an indication of how to plan and implement experiments with effective results.

2 Literature Review About a decade ago Virtual Reality and Augmented Reality had already proven effective in manufacturing. With the applicability of Augmented Reality mainly in the production support and maintenance processes but also with applications of Virtual Reality for design, training, prototyping and marketing processes. The Fourth Industrial Revolution started to be previewed a decade ago (Culot et al. 2020) [7] today we are witnessing the beginning of it (…) This revolution is characterized by a widespread application of the Internet which has become easier to use, and the application of intelligent components, robots and technologies (Damiani et al. 2018) [9]. Supporting the creation of the “smart factory” in which physical and digital systems are integrated with the aim of reaching “mass personalization” and “faster product development” (Demartini et al. 2017) [11], each entity from the physical world has its digital twin in the digital world and so the entire production system can be simulated. These systems are characterized by context awareness, which allows people and machines to perform their tasks in an optimal manner (Lucke et al. 2008) [15]. The great objective of the last decade is now taking the first steps with greater precision due to the great technological advance of equipment, allowing its use with greater ease, making it less uncomfortable and with less disturbance for users during long periods of time. Despite these great advances, these tools, such as Nee and Ong (2013) [17] refer, still face a higher order of accuracy, response, and interaction design. The head-mounted display (HMD) devices have been a popular choice when AR applications were first developed, as the eye-level display facilitates direct perception of the combined AR scene. HMD devices, however, are uncomfortable and may cause headache and dizziness, especially after prolonged usage.

14

A. Neto et al.

Despite the great strides made in this period, at present we are still waiting for improvements, according to Xiong (Xiong et al. 2021) [26], in terms of display performance, AR and VR face several common challenges to satisfy demanding human vision requirements, including field of view (FoV), eyebox, angular resolution, dynamic range, and correct depth cue, etc. Another pressing demand, although not directly related to optical performance, is ergonomics. To provide a user-friendly wearing experience, AR and VR gear should be lightweight and ideally have a compact, glasses-like form factor. These authors also present a prediction that the next generation will be focused on AR and VR technologies as display platforms for deeper human-digital interactions due to the rapid advances we are seeing in high-speed performance computing and communication. We are currently witnessing the increasingly visible start of the presence of these technologies in manufacturing processes with many proven advantages, with greater emphasis on the Augmented Reality technology (Nee and Ong) [17] - AR-assisted simulation tools that could improve manufacturing operations, as well as product and process development, leading to faster learning and training, hence shorter lead-time, and consequently, reduced cost and improved quality. Virtual Reality technology has become common in industry and has gained in cost competitiveness. Thus, it is now broadly recognized that it is valuable for manufacturing companies to invest in VR. It is a helpful technology in achieving rapid understanding and decision-making by visualization and experience, applying not only to product design but also to production actively. It can be used as a dashboard for key performance indicator (KPI) monitoring on the shop floor. For this, the development of dynamic connections and integration between the element technologies of manufacturing, VR technologies, and IT technologies are important. (Choi et al. 2015) [6]. Augmented Reality technology is already used in many industrial environments to help address product lifecycle challenges, including planning, design, ergonomics assessment, maintenance, management and training (Wang et al. 2016) [24]. The following future research areas have been singled out as the most promising: object recognition and position; eye and voice control; the control of industrial systems via AR. These areas allow the combination of existing technologies and the exploitation of their best characteristics in order to obtain the highest achievements and thus improve the digital industrial environment (Reljic´ et al. 2021) [18].

3 Technology Overview In this chapter the immersive technologies named Virtual Reality and Augmented Reality will be described and defined in their origins and evolution from the early beginning until the present days. 3.1 An Overview About Virtual Reality Virtual Reality is a technology linking a user and an operating system through threedimensional (3D) graphics or 360° images whose objective is to create the sensation of presence in a virtual environment different from the real one.

How Virtual and Augmented Reality Can Boost Manufacturing

15

With the use of equipment known as Virtual Reality glasses (or even headset), the interaction provides a feeling of being present in a virtual environment, completely disconnected from the environment in which the person is physically. This sense of presence is usually referred to as virtual immersion. There is an even simpler definition: Virtual Reality is the use of technology to convince the user that he is in another reality. 3.1.1 Origin of Virtual Reality One of the first attempts at Virtual Reality is believed to be stereoscopes, whose most rudimentary form was invented in 1838 by physicist Sir Charles Weatstone (Varnum 2019) [23]. One of these first creations, along the lines of the Virtual Reality we know today, was Sensorama, invented by Morton Heilig in 1956 (Heiling, 1962) [12]. Sensorama was a simulator with a 3D screen, stereo sound, body inclination, and sensations such as wind and aromas, that projected the experience of a motorcycle running through Brooklyn, New York, U.S.A (see Fig. 1). A few years later, in 1965, Ivan Sutherland wrote an essay under the title “The Ultimate Display” which resulted in the construction of a rudimentary prototype of this device, which he called the “Sword of Damocles” (Sutherland, 1965) [21]. This device is considered the first head-mounted display (HMD) system for Virtual Reality and Augmented Reality. Since 1960, Thomas Furness had been working on displays and instrumentation for cockpits in the US Air Force. In the late 1970s, he began to develop visual interfaces for aircraft control and in 1982 he introduced the “Visually Coupled Airborne Systems Simulator” (VCASS) (Kocian, 1977) [14] also known as the “Darth Vader helmet”. Between 1985 and 1989, Thomas Furness directed the US Air Force’s “Super Cockpit” program. Using HMD, he developed a system capable of projecting information such as 3D maps, radar and data into a 3D virtual space that the pilot could see and hear in real time (Curry, 1985) [8].

Fig. 1. Sensorama

16

A. Neto et al.

The HMD’s motion detection system, voice controls and other sensors allowed the pilot to control the plane with speech, gestures and eye movements (Cipresso, et al., 2018) [5]. 3.1.2 Evolution of Virtual Reality In terms of Virtual Reality systems we can find three types: • Non-immersive systems are the simplest and cheapest type of VR applications that use desktops to reproduce images of the world (Cipresso et al., 2018) [5] to reproduce virtual environments mainly with the use of 360º images or videos. • Immersive systems provide a complete simulated experience due to the support of several sensory output devices such as head-mounted displays (HMDs) for enhancing the stereoscopic view of the environment through the movement of the user’s head, as well as audio and haptic devices [5]. • Semi-immersive systems such as Fish Tank VR are between the two above. They provide a stereo image of a three dimensional (3D) scene viewed on a monitor using a perspective projection coupled to the head position of the observer (Ware et al., 1993) [25]. Adding to these systems, in order to provide a higher immersion, it is possible to add other technologies that allow activating more human senses, such as smell, touch, among others. According to Burdea (Burdea et al., 1996) [2] we can distinguish Virtual Reality technologies by input devices and output devices. In short, input devices are the ones that allow the user to communicate with the virtual environment, which can range from a simple joystick or keyboard to a glove allowing capturing finger movements or a tracker able to capture postures. Output devices allow the user to see, hear, smell, or touch everything that happens in the virtual environment. Following we can see a set of Virtual Reality equipment that demonstrates the evolution of the VR devices in the last decade. Some of them use mobile devices as processor and display, other ones use Personal Computers (PC) as base for displaying the experiences.

Fig. 2. Google Cardboard and Daydream [27]

Launched in 2014, Cardboard is a VR platform developed by Google (see Fig. 2). Named after its format, the platform was designed as a low-cost system to stimulate

How Virtual and Augmented Reality Can Boost Manufacturing

17

interest and development in VR applications. When using mobile devices to play content, cardboard lenses allow the viewer to experience an immersive experience anywhere. This platform was one of the biggest developments for immersive technologies as it allowed anyone to experiment quickly and cheaply. Along with this diverse cardboard, models like Google Daydream (see Fig. 2) were developed with a different format, more comfortable, but using the same technology.

Fig. 3. Samsung Gear VR [28]

Fig. 4. VR Box

Samsung Gear VR (see Fig. 3) and VR Box (see Fig. 4) are examples of the evolution of the Google Cardboard platform. The VR Box works exactly in the same way as the cardboard. Users need to insert a mobile device and the headset lenses will provide the immersive experience. The evolution in this case is the remote controller that connected via bluetooth with the mobile device allowed the user to interact in multiple ways with the experience. Also released, also in 2014, the Samsung Gear VR platform which was an important improvement in the quality of the immersive experiences. Using high quality display mobiles, this platform allowed developers to create high quality experiences that could be seen in VR using Samsung´s mobiles.

Fig. 5. Oculus DK1 and DK2 [29]

An American startup named Oculus, in 2012 was researching how they could build VR equipment, came out with the DK1 (Development Kit 1) and DK2 (Development Kit 2) - PC-based headsets (see Fig. 5). Those were the first versions of what would be the beginning of the new era of VR headsets. Oculus startup also created a Kickstarter campaign to fund the development of the next generation of their VR headsets, the Rift series (see Fig. 6), and it was successfully funded, raising ten times the original goal. In March 2014, Meta, formerly known as Facebook, acquired the startup and started a new chapter in the VR headset industry.

18

A. Neto et al.

Fig. 6. Oculus Rift [31]

Fig. 7. HTC Vive [32]

HTC Vive headsets (see Fig. 7) were released in 2015/2016 and it plays an important role in this industry. Vive headsets had high-quality displays with the highest display resolution of the market and with an important development of sensors, the “lighthouses” that improved the motion capture of the users. Being those sensors installed in the wall, they could capture the motion of the users, translating this into more faithful recognition of their movements, making the experiences even more immersive for the users.

Fig. 8. Sony PlayStation VR [33]

Fig. 9. Oculus Go [30]

Sony Playstation VR (see Fig. 8) is known as one of the biggest players in the VR market. The VR headsets are available for PlayStation game consoles and it was released in 2016 and since Sony PlayStation is one of the most popular brands in video games, the VR headset for PlayStation got popular as well. Sony officially announced that they have reached its 5 million unit sales milestone in 2020 [33]. Released in 2017, the Oculus Go (see Fig. 9) is a standalone Virtual Reality headset developed by Meta (formerly known as Facebook) in partnership with Qualcomm and Xiaomi. It was the first generation of Oculus Virtual Reality headsets, in the category of standalone VR headsets. Oculus Quest (see Fig. 10) was the natural evolution of Oculus Go. It is a standalone (wireless) device that can run games and software wirelessly under an Android-based operating system with two controllers. It supports positional tracking with six degrees of freedom (6DoF), using internal sensors and cameras, located in the front of the headset, also used as part of the safety feature called “passthrough”, which shows a view from the cameras. Quest received a warm welcome from the market for its price and convenience, and for having improved graphical fidelity and tracking over Oculus Go.

How Virtual and Augmented Reality Can Boost Manufacturing

Fig. 10. Oculus Quest [34]

19

Fig. 11. PRIMAX 8k [35]

Pimax Technology, founded in 2015, has developed a series of VR headsets with high-quality displays. Its first product, the Pimax 4K (see Fig. 11), was launched in 2016, making it the first 4K headset available on the market. In 2017, they ran a crowdfunding campaign for an even better headset, the Pimax 8K, raising approximately $4.2 million, which held the Guinness World Record for the most successful crowdfunding VR project to date.

Fig. 12. Valve Index [36]

Fig. 13. HP Reverb G2 [37]

Valve Index is a VR headset developed by Valve (see Fig. 12). Launched in 2019, the Index is a second-generation Valve headset, but the first to be manufactured entirely by them. It features full RGB screens with high-resolution, 130° of FoV (field of view) and controllers that are compatible with Valve’s and HTC (Lighthouse) tracking systems. Developed by Hewlett Packard in collaboration with Valve and Microsoft, HP Reverb (see Fig. 13) delivers a more immersive, comfortable, and compatible experience with improved tracking below the waist with more vertical area coverage, and an adjustable eye relief facemask to customize eye distance from the lenses and get an improved visual experience. Equipped with industry-leading lenses and speakers this HMD provides high-quality resolution and fully immersive spatial audio. It is the successor to Facebook’s previous headset, the Oculus Quest. Officially unveiled on September 16, 2020 during the Facebook Connect 7 event in the United States of America.

20

A. Neto et al.

Fig. 14. Oculus Quest 2 [38]

Fig. 15. Varjo [39]

The 64 GB launch model was priced at $299, a $100 reduction from the original Oculus Quest. In 2021, the 64 GB Quest 2 model was replaced by the 128GB model, keeping the same price. Like its predecessor, the Quest 2 (see Fig. 14) is capable of functioning as a standalone headset with an Android-based internal operating system and Oculus-compatible VR software running on a desktop computer when connected via USB or Wi-Fi. It is an update of the original Oculus Quest with a similar design but lighter weight, updated internal specs, a display with a higher refresh rate and resolution per eye, and updated Oculus Touch controllers. Varjo is a Finnish manufacturer of Virtual Reality, Augmented Reality and Mixed Reality headsets (see Fig. 15). Founded in 2016 by former Nokia and Microsoft executives, Varjo specializes in developing high-resolution devices. 3.2 An Overview About Augmented Reality Augmented Reality is the integration of elements or virtual information with real-world views through a camera and with the use of motion sensors such as a gyroscope and accelerometer. A person experiencing Augmented Reality can use translucent glasses (headsets) or cameras coupled to a mobile device (mobile phones or tablets), and through these, can see the real world as well as computer generated images projected in their field of vision. 3.2.1 Origin of Augmented Reality Ivan Sutherland developed the first Head Mounted Display or HMD in 1968 (see Fig. 16), which he called “The Sword of Damocles” [21]. It was a head-worn device with an optical display in front of one (HMD Monocular) or each eye (HMD Binocular) for viewing 3D objects in the real environment and attached to the ceiling of his laboratory due to its weight. Although the concept has been around for a long time, the term Augmented Reality was only adopted in 1992 by researcher Thomas Caudell, a Boeing engineer [4]. Thomas Caudell and David Mizell were challenged to present an alternative to the diagrams and marking devices used to guide workers on the company’s shop floor during aircraft production.

How Virtual and Augmented Reality Can Boost Manufacturing

21

The solution they developed was an HMD that could be used by workers and displayed the diagrams and schematics of the planes and projected them on reusable boards, and the information to be visualised could be changed using a computer.

Fig. 16. Demonstração do primeiro HMD pelo Ivan Sutherland

3.2.2 Evolution of Augmented Reality In terms of Augmented Reality technologies, although there is a lot of diversity (Carmigniani et al., 2011) [3] presented three common components, such as a geospatial datum for the virtual object, like a visual marker, a surface to project virtual elements to the user, and an adequate processing power for graphics, animation, and merging of images, like a pc and a monitor. To run an AR system it is necessary to include a camera able to track the user movement for merging the virtual objects, and a visual display, like glasses through which the user can see the virtual objects overlapping to the physical world. Following we can see a set of Augmented Reality equipment that demonstrates the evolution of the AR devices in the last decade. AR experiences can be seen using mobiles and tablets using Android and iOS operational systems (see Fig. 17). Some experiences require more powerful mobiles and tablets for image and sound processing and most of the recently launched mobiles can provide that. There are two important sensors required for AR experiences in mobile devices: Accelerometer and Gyroscope. Those sensors are responsible to report to the application the position of the mobile allowing the interaction with the user. Mira Prism Pro is a quite affordable (see Fig. 18), lightweight AR headset that uses mobile devices inside the headsets to mirror the content into the lenses. Thanks to giant lenses, Mira Prism provides a better field of view than most of the other smaller AR glasses.

22

A. Neto et al.

Fig. 17. Mobile telephones and tablets used for Augmented Reality experiences

Fig. 18. Mira Prism Pro [46]

Fig. 19. Lenovo ThinkReality [68]

Lenovo SmartGlasses (see Fig. 19) offers enterprise-grade AR solutions that change the way people work in the field, allowing them to receive assistance, streamline complex workflows and improve training quality easily.

Fig. 20. Epson Moverio [69]

Fig. 21. Vuzix Blade [70]

Binocular glasses with projection technology for better images with integrated sensors and enhanced connectivity features, seamlessly blend digital content into the outside world. Moverio’s lightweight, flexible, hinge-based headset designs accommodate a wide range of head sizes and make wearing comfortable, even for extended periods of time (see Fig. 20). Vuzix Blade is a high-level, lightweight AR glasses (see Fig. 21) that enables seamless remote collaboration and workforce digitization. The device delivers a handsfree connection of the digital world to reality, offering users high-value access to location-aware information, data collection, and full UV protection lenses.

How Virtual and Augmented Reality Can Boost Manufacturing

23

Fig. 22. RealWear Navigator 500 [71]

Navigator 500 was designed as a modular platform with an upgradeable 48megapixel camera system, a hot-swappable battery, Wi-Fi, and an optional 4G modem. The voice-controlled user interface includes unique noise-cancellation technology designed for high-noise environments. RealWear smart glasses are Assisted Reality ones (see Fig. 22). Assisted Reality technology differs from Augmented Reality in a key way. Assisted Reality gives users access to relevant information in their immediate field of view (FoV), while Augmented Reality uses computer-generated, overlaid digital content to create an interactive experience within real-world environments.

Fig. 23. Microsoft Hololens 1 and 2 [41]

An ergonomic, untethered self-contained holographic device with enterprise-ready applications to increase user accuracy and output, the Microsoft Hololens 1 and 2 (see Fig. 23). Those headsets are part of what is called Mixed Reality (MR). In a brief definition, MR is the merging of real and virtual worlds that produces new environments and visualizations, where physical and virtual objects co-exist and interact in real-time.

Fig. 24. Magic Leap [72]

24

A. Neto et al.

Heads-up display is equipped with sophisticated sensors and world-understanding, capturing the contour and content of your workspace so that applications intelligently integrate into your environment. Magic Leap (see Fig. 24) is part of what is called Extended Reality (XR). Briefly defined, it is a term that refers to all combined real and virtual environments and humanmachine interactions generated by computer technology and wearables. It includes Augmented Reality, Mixed Reality (RM), Virtual Reality and the areas interpolated between them.

Fig. 25. N Real [73]

N Real’s Light sunglasses (see Fig. 25) are one of only a few consumer-focused augmented reality headsets. It is also an XR-based smart glasses. 3.2.3 Use Cases - Known Use Cases Using Immersive Technologies Below, we can analyze some use cases of the technologies explained so far. Most of them are applications of Augmented Reality technology in our real life or brands related to our daily tasks. Although some of them are not related to industrial application, it is interesting to understand the possibilities that this technology provides.

Fig. 26. Nike Air [74]

Fig. 27. Burger King – Burn that ad [75]

A quite interesting example, Nike Air app (see Fig. 26), of the use of AR for brands where the customer at home can try new shoes even before buying it. An example of how retail can use AR for Marketing in a fun way - Burger King “Burn that ad” (see Fig. 27).

How Virtual and Augmented Reality Can Boost Manufacturing

Fig. 28. IKEA – AR Furniture (_)

25

Fig. 29. Google Maps (_)

IKEA developed an AR application and clients could try new furniture without leaving home, with real simulation (see Fig. 28). Navigation apps (see Fig. 29) are learning to use your camera to locate you far better than GPS ever could—and give you better directions to wherever you’re headed.

Fig. 30. National Geographic – AR [79]

Fig. 31. National Geographic – VR [80]

National Geographic provided a quite interesting experience using AR in a giant mirror in a train station in Rotterdam (see Fig. 30). Another National Geographic development, this time using 360º videos to demonstrate how lions interacted with the natural environment (see Fig. 31). This content can be viewed using VR headsets or standart PC.

4 Industrial Application Finished the overview about the definition of Virtual Reality and Augmented Reality, origins, technologies, evolution and known use cases using immersive technologies, now it is time to frame it in the industrial area - specifically in manufacturing. It will be exposed the five main areas in industrial application, also suggestions on how to implement it, and finally, examples of tools using Augmented or Virtual Reality. Application of Immersive Technologies in Manufacturing The new paradigm known as the “fourth industrial revolution” or “Industry 4.0” was first used in 2011 to cover two different meanings: as a synonym for the new industrial revolution, triggered by steam-powered mechanization, electricity and information, and

26

A. Neto et al.

communication technologies (ICT); also as a label for the strategic plan pursued by Germany to strengthen its international competitive position in manufacturing (Kagermann et al., 2013) [13]. As mentioned in Culot article (Culot et al. 2020) [7], despite the origins of this term being seldom questioned, at the same period many other places in the world were working on the idea of a “manufacturing renaissance”: government initiatives, academic studies and project reports. Other concepts emerged speaking of the same phenomenon - “Industrial Revolution”, “Smart Manufacturing”, “Smart Factory”, “Industrial Internet”, among others, prevailing the term “Industry 4.0”. This term refers to, according to Damiani (Damiani et al., 2018) [10], new production patterns, including new technologies, productive factors and labour organizations, which today we are witnessing that are completely changing the production processes and the relationship between customer and company with relevant effects on the supply and value chains. One of the studies that had a great impact on today’s Industry 4.0, was defined by NASA in 2010, the “digital twins” as an integrated multi-physics, multi-scale, probabilistic simulation of a vehicle or a system that uses the best available physical models, sensor updates, fleet history, etc., to mirror the life of its corresponding flying twin (Shafto et al. 2010) [19]. According to Trainer (Trainer et al., 2020) [22], it exists a lot of discussions around the definition of digital twins, but in their article they highlight a more current definition by Stark and Damerau (2019) [20], in manufacturing defines the term as a digital representation of an active unique product […] or unique product-service system […] that comprises its selected characteristics, properties, conditions, and behaviours by means of models, information, and data within a single or even across multiple life cycle phases. Virtual and Augmented Reality technologies in manufacturing have a set of advantages [9]: i) speed up reconfiguration of production lines, ii) support shop-floor operators, iii) implement virtual training for assembling parts, iv) manage the warehouse efficiently, v) support advanced diagnostics integrated into the modules and vi) interact with the working environment minimizing risk. Next it will be presented five main areas in industrial application, with examples of best practices to better understand their functionality. 4.1 Five Main Areas in Industrial Application There are five main areas in industrial application that have a lot of results and proven advantages: Simulation and Training; Remote Field Services; Team Collaboration; Sales and Marketing and Product or Project Design. It is important to emphasize that these five chosen areas are directly related to faster investment returns. Despite being applied in specific areas in industries, usually, the implementation of projects like this, demands the integration of one more team in the companies. As an example, we can cite a training experience that can be applied in one production area but requires IT involvement to enable the team with equipment, from HR in the construction of the best practices, content and training metrics, in the related area for availability of the content and effective training of new employees and also Marketing to present the content and results within the company.

How Virtual and Augmented Reality Can Boost Manufacturing

27

Some of the existing use cases show return on investment in the second or third cycle of use despite the high investment required to develop the experiments. Next, each of these areas will be described showing the advantages of each for the manufacturing industry. 4.1.1 Simulation and Training In the areas of Simulation and Training, an improvement in employee learning process and performance can be observed, as the use of VR allows a greater focus on the content presented to the user. This focus also allows an increase in the speed of learning cycles, with a reduction in the costs and makes it easier to train remote teams and also employees who need special formats with accessibility. These immersive training experiences allow the development of more engaging multimedia content that will keep employees more committed and, consequently, successful in completing training cycles. Interestingly, once the training experiences are created, the repetition or creation of new cycles becomes much easier and faster and at a lower cost. The same goes for updating the content. Next, a good practice will be exemplified (see Fig. 32).

Fig. 32. Simulation and training - Virtual Reality Motor Maintenance by Mechatraining LLC [40]

4.1.2 Remote Field Services In Field Services, the application of immersive technologies such as Augmented Reality, helps technicians in the field to boost first-time fix rates. With this, an increase in the productivity of technicians can be seen, which can also reduce operational errors as they are using collaborative tools with specialists in the back office. AR techniques also allow for an improvement in the training of new employees as it is possible to present content in real-time and not in books or videos anymore. But one of the main benefits, which is not much observed, is the reduction in equipment downtime. Using immersive technologies, technicians can start solving the problem in minutes rather than days until a technician checks into the field.

28

A. Neto et al.

In summary, remote field services: boosts first time fix; increases technician productivity; reduces technician error; allows remote expert assistance; improves and shortens novice training and reduces costly downtime. Next, a good practice will be exemplified (see Fig. 33).

Fig. 33. Remote field services - Remote Assist for Microsoft HoloLens 2 [41]

4.1.3 Team Collaboration When we talk about team collaboration, especially when you have teams in different places, VR and AR collaborative tools greatly increase productivity and focus during sessions. Imagine that during a virtual meeting, when wearing VR glasses, employees will be more focused on the content presented, as they will be immersed in the content, without any distractions. In addition, the use of multimedia resources, allow the collaborative sessions and even training, to be richer with greater ease in presenting content and learning new theories. It is also possible to use collaborative tools to better manage corporate culture with teams in different countries by presenting the same content in an immersive way, with great creativity from the creative teams and creating better corporate communication channels. In summary, team collaboration: provides better focus during collaboration; enhanced training opportunities; drives better company culture; greater creativity and improved communication. Next, a good practice will be exemplified (see Fig. 34). 4.1.4 Sales and Marketing Sales and Marketing are the most common areas of VR and AR applications across all markets. Retail uses AR in product sales campaigns in several countries around the world and the same can be used in industries. The presentation of equipment on sites, even before customers demand it, allows the analysis of application, space organization, adaptation to the production model and visual testing of this equipment.

How Virtual and Augmented Reality Can Boost Manufacturing

29

Fig. 34. Team collaboration - Microsoft Mesh [42]

AR and VR presentations are great tools for commercial teams at trade shows and events to present content before they even exist by creating showrooms anywhere. Virtual stores or pop-up stores can be created in any physical space using tablets and mobile phones to present products and services, directly from the seller’s pocket. In summary, in the sales and marketing area the advantages are: being one step ahead of your competitors; you can showcase your services to potential customers in advance; VR offers ec + nhanced user experience; XR marketing helps boost sales; it’s a perfect tool for content marketing. Next, a good practice will be exemplified (see Fig. 35).

Fig. 35. Sales and Marketing - Virtual Advertising - Mercedez Benz [43]

4.1.5 Product or Project Design Another interesting industrial application is the area of project and product development using VR and AR visualization. Until some time ago, it was only thought of using these tools individually, such as in CAD or 3D tools, but nowadays you can see the collaboration between designers who are in different places and developing the same project, in real time using headsets and controls. We can talk about car design, architectural projects, engineering projects and get the final result virtually without wasting materials. All prototypes can be made virtually and tweaked and tested with a smaller amount of time, risk and money. The power of viewing content in a virtual and immersive way is impressive.

30

A. Neto et al.

In summary, in product or project design, the advantages are: seeing versus predicting - the designer can see the end result of their work before it has been built; better space visualisation; massive savings during prototyping and allows team collaboration. Next, a good practice will be exemplified (see Fig. 36).

Fig. 36. Product or Project Design - McLaren Automotive [44]

Some other applications of immersive technologies that mix the previous areas of industry application, can be seen in the following images: AR - PTC Vuforia (see Fig. 37), VR - Matteport (see Fig. 38), VR - 3 Data (see Fig. 39), AR/VR Virtualics (see Fig. 40), VR - Bublar Group and Dafo (see Fig. 41).

Fig. 37. Field Service and Training with AR - PTC Vuforia [45]

4.2 Suggestions on How to Implement Augmented or Virtual Reality Projects The development of experiences in VR or AR requires preparation even before hiring teams to develop your experience. And that is why it is important to have these five steps prepared for hiring a professional or a company for development: 1- Define your goals: What is the expected objective using the experience? What is your expectation from this? Set goals, and KPIs (Key Performance Indicator) that can be evaluated in the very first test. Thus, it is possible to measure the effectiveness of using this new tool with your team and it will be possible to justify the cost involved.

How Virtual and Augmented Reality Can Boost Manufacturing

31

Fig. 38. Training and Sales with VR - Matterport [46]

Fig. 39. Engineering and Design with VR - 3Data [47]

Fig. 40. Data Visualisation with AR or VR - Virtualics [48]

2- Define the target audience: It is important to understand, even before developing the experience, if your users are able to use the technology, if it is easy to use and if they will have guidance on the use of the equipment. Remember that in the implementation phase, it is necessary to follow up on experiences and feedback is very important to improve the content and experience.

32

A. Neto et al.

Fig. 41. Fire emergency training solution from Bublar Group [49]

3- Develop the content: Before even developing the experience, develop with your team what should be presented, how it should be presented and think of all the ways this can be done so that developers have alternatives in the presentation of the content. 4- Approve the budget: The development budget can change throughout the development phase, but if the planning is done well, margins can minimize these fluctuations. Always be aware that when you change the content or scope, you change the cost. That’s why good planning supports a good budget. 5- Search for a developer or service provider: After completing the previous steps, do market research on the type of service you are going to hire. If you want to hire employees for in-house development, if you want to hire third parties who can develop a specific solution or even hire a studio/company to develop the solution. These five steps should be followed so that you don’t have any surprises with the final result of the experience and even so, it is necessary to follow up point-by-point of the deliverables and milestones of the development project. Follow those steps and you will get the jail-breaking “wow-factor” from your colleagues or clients that are using XR for the first time. Enjoy this moment and take advantage of it to ask for funding for your next project. 4.3 Example of Tools Developed Using Augmented or Virtual Reality Taking into account all the topics presented so far, here are references to tools and startups that develop content related to each topic. 4.3.1 Simulation and Training – Kit-AR - Enables manufacturers to continuously improve production quality, by enabling shopfloor teams with tech tools that reduce errors and waste. [50] – Glartek - software created to increase the efficiency and safety of industrial tasks performed by frontline workers. [51] – Immerse - companies create, deploy, scale, measure and integrate VR training and maximise their ROI. [52]

How Virtual and Augmented Reality Can Boost Manufacturing

33

– Reflekt - Remote Work with Augmented Reality and live video on mobile devices and wearables for front-line workers and technicians. [53] – Pixo VR - Get the Extended Reality content you need, then distribute and manage it simply and securely on a global scale. [54] 4.3.2 Team Collaboration – Glue - A virtual collaboration platform for teams who need remote meetings to be as great as face-to-face meetings. [55] – Spatial - Spatial is dedicated to helping creators and brands build their own spaces in the metaverse to share culture together. [56] – Microsoft Mesh - enables presence and shared experiences from anywhere – on any device – through mixed reality applications. [57] – Arthur - Enterprise collaboration in Virtual Reality. Arthur enables your organization to meet, collaborate and manage work. [58] 4.3.3 Remote Field Services – SightCall - SightCall is an enterprise-grade video cloud platform helping service leaders improve outcomes without deploying unnecessary support to the field. [59] – LibreStream - Transform your workforce with Librestream and enhance safety, efficiency and resiliency. [60] – TeamViewer - Cloud-based platform with intuitive features to securely and remotely access, control and support any device, across platforms—from anywhere, anytime. [61] 4.3.4 Sales and Marketing – Matterport - Take your buildings online with Matterport to design, build, promote, and manage your most valuable asset at your fingertips. [46] – 8th Wall - Use 8th Wall’s powerful platform to develop augmented reality that works on all devices, no app required. [62] – BlippAR - Push boundaries and make augmented reality that’s valuable and entertaining. [64] 4.3.5 Product or Project Design – The Wild - Discover the best way for architects & building teams to charrette, collaborate, and present designs at human-scale—from anywhere. [65] – IrisVR - Load your 3D file and click launch—that’s it. The VR menu is built for presentations, collaborative sessions, and design review. [66] – Unity Reflect Review - Communicate effectively during design reviews, facilitate more efficient collaboration with project stakeholders, avoid costly mistakes, and drive alignment faster. [67]

34

A. Neto et al.

5 Conclusions The digital transformation process follows a one-way road. Industries need to adapt to the new production model where all processes should be digitized and immersive technologies can support that. We are talking about the construction of digital twins about customer experience but mainly about digital strategies. It is necessary to change that old mindset that only core processes need to be digitized. All peripheral and dependent processes on production lines can and should be digitized in search of process optimization, with fairer costs and better use of resources. It is also important to remember about communities related to the development of those new technologies. It is always through communities that recommendations are obtained about best practices, hardware, software and the development of use cases. Communities are often informal and can be supported by industry or academia and it is always important to be a part of and support these initiatives.

References 1. Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S., MacIntyre, B.: Recent advances in augmented reality. IEEE Comp. Graph. Appl. 21, 34–47 (2001). https://doi.org/10.1109/ 38.963459 2. Burdea, G., Richard, P., Coiffet, P.: Multimodal virtual reality: input-output devices, system integration, and human factors. Int. J. Hum. Compu. Interact. 8, 5–24 (1996). https://doi.org/ 10.1080/10447319609526138 3. Carmigniani, J., Furht, B., Anisetti, M., Ceravolo, P., Damiani, E., Ivkovic, M.: Augmented reality technologies, systems and applications. Multimed. Tools Appl. 51, 341–377 (2011). https://doi.org/10.1007/s11042-010-0660-6 4. Caudell, T.P., Mizell, D.: Augmented reality: an application of heads-up display technology to manual manufacturing processes. In: System Sciences, 1992. Proceedings of the TwentyFifth Hawaii International Conference on Volume: ii (1992). https://doi.org/10.1109/HICSS. 1992.183317 5. Cipresso, P., Giglioli, I.A.C., Raya, M.A., Riva, G.: The past, present, and future of virtual and augmented reality research: a network and cluster analysis of the literature. Front. Psychol. (2018). https://doi.org/10.3389/fpsyg.2018.02086 6. Choi, S., Jung, K., Not, S.D.: Virtual reality applications in manufacturing industries: past research, present findings, and future directions. Concurrent Eng. Res. Appl. (2015). DOI:https://doi.org/10.1177/1063293X14568814 7. Culot, G., Nassimbeni, G., Orzes, G., Sartor, M.: Behind the definition of Industry 4.0: analysis and open questions. Int. J. Prod. Econ. (2020). https://doi.org/10.1016/j.ijpe.2020.107617 8. Curry, D.: Cockpits of the future. Airman-official Magazine of the U.S. Air Force. In: Betezel, A. (ed.) (Np-10) October 1985, vol. XXIX, pp. 9–13 (1985) 9. Damiani, L., Demartini, M., Guzzi, G., Revetria, R., Tonelli, F.: Augmented and virtual reality applications in industrial systems: a qualitative review towards the industry 4.0 era. IFAC (International Federation of Automatic Control), PapersOnline 51–11. Elsevier, Napoli (2018) 10. Damiani, L., Revetria, R., Volpe, A.: Augmented reality and simulation over distributed platforms to support workers’. In: 2015 Winter Simulation Conference (WSC), pp. 3214–3215 (2015). https://doi.org/10.1109/WSC.2015.7408476

How Virtual and Augmented Reality Can Boost Manufacturing

35

11. Demartini, M., Toneli, F., Damiani, L., Revetria, R.: Digitalization of the manufacturing execution systems: the core technology for realizing future smart factories. In: XXII Summer School “Francesco Turco” - Industrial Systems Engineering. Dalmine, Italy (2017). http://www.summerschool-aidi.it/edition-2017/cms/extra/papers/47-%20Demartiniwith-numbers.pdf. Accessed 15 Nov 2021 12. Heilig, M.: Sensorama simulator. U.S. Patent No - 3, 870. United States Patent and Trade Office, Virginia (1962) 13. Kagermann, H., Wahlster, W., Helbig, J.: Recommendations for implementing the strategic initiative Industrie 4.0. Final report of the Industrie 4.0 Working Group. National Academy of Science and Engineering (2013). http://www.acatech.de/fileadmin/user_upload/Baumst ruktur_nach_Website/Acatech/root/de/Material_fuer_Sonderseiten/Industrie_4.0/Final_rep ort__Industrie_4.0_accessible.pdf 14. Kocian, D.F.: A visually-coupled airborne systems simulator (VCASS) - an approach to visual simulation. In: Conference Paper - Air Force Aerospace Medical Research Lab WrightPatterson AFB OH. Defense Technical Information Center. Virginia, USA (1977) 15. Lucke, D., Constantinescu, C., Westkämper, E.: Smart factory - a step towards the next generation of manufacturing. In: Mitsuishi, M., Ueda, K., Kimura, F. (eds.) Manufacturing Systems and Technologies for the New Frontier, pp. 115–118. Springer, London (2008). https://doi.org/10.1007/978-1-84800-267-8_23 16. Milgram, P., Kishino, F.: A taxonomy of mixed reality visual displays. IEICE Trans. Inform. Syst. 77, 1321–1329 (1994) 17. Nee, A.Y.C., Ong, S.K.: Virtual and augmented reality applications in manufacturing. In: 7th IFAC Conference on Manufacturing Modelling, Management, and Control. Saint Petersburg, Russia (2013) 18. Relji´c, V., Milenkovi´c, I., Dudi´c, S., Šulc, J., Bajˇci, B.: Augmented Reality Applications in Industry 4.0 Environment. MDPI, Basel. https://doi.org/ https://doi.org/10.3390/app111 25592. (2021) 19. Shafto, M., et al.: DRAFT modeling, simulation, information technology & processing roadmap. Technol. Area 11 (2010). https://www.nasa.gov/pdf/501321main_TA11-MSITPDRAFT-Nov2010-A1.pdf. Accessed 15 Nov 2021 20. Stark, R., Damerau, T.: Digital twin. In: Chatti, S., Tolio, T. (eds.) CIRP Encyclopedia of Production Engineering, vol. 66, pp. 1–8. Springer, Heidelberg (2019). https://doi.org/10. 1007/978-3-642-35950-7_16870-1 21. Sutherland, I.E.: The Ultimate Display. Multimedia: From Wagner to Virtual Reality. New York (1965). http://worrydream.com/refs/Sutherland%20-%20The%20Ultimate%20D isplay.pdf. Accessed 15 Nov 2021 22. Trainer, J., Schweigert-Recksiek, S., Spreitzer, K., Zimmermann, M.: What is a digital twin? Definitions and insights from an industrial case study in technical product development, conception of digital twins for technical product development. Technische Universität München. (2020). https://doi.org/10.1017/dsd.2020.15 23. Varnum, J.K.: Beyond Reality - Augmented, Virtual and Mixed Reality in the Library, p. 3. Ala Editions, Chicago (2019) 24. Wang, X., Ong, S.K., Nee, A.Y.C.: A comprehensive survey of augmented reality assembly research. Adv. Manuf. 4(1), 1–22 (2016). https://doi.org/10.1007/s40436-015-0131-4 25. Ware, C., Arthur, K., Booth, K.S.: Fish tank virtual reality. In: Proceedings of the INTERACT 1993 and CHI 1993 Conference on Human Factors in Computing Systems, pp. 37–42. ACM, Amsterdam (1993). https://doi.org/10.1145/169059.169066 26. Xiong, J., Hsiang, E., He, Z., Zhan, T., Wu, S.: Augmented reality and virtual reality displays: emerging technologies and future perspectives. Nat. Official J. CIOMP 2047-7538. Orlando, USA (2021)

36

A. Neto et al.

27. Google AR & VR. https://arvr.google.com/. Accessed 15 Nov 2021 28. Samsung Gear VR. https://www.samsung.com/global/galaxy/gear-vr/. Accessed 15 Nov 2021 29. Oculus DK1 and DK2. https://www.oculus.com/blog/update-on-developer-kit-technologyshipping-details/. Accessed 15 Nov 2021 30. Oculus Go. https://www.oculus.com/go. Accessed 15 Nov 2021 31. Oculus Rift. https://www.oculus.com/rift-s/. Accessed 15 Nov 2021 32. HTC Vive. https://www.vive.com/eu/. Accessed 15 Nov 2021 33. Sony Playstation VR. https://www.sie.com/en/corporate/release/2020/200107.html. Accessed 15 Nov 2021 34. Oculus Quest. https://www.oculus.com/quest/. Accessed 15 Nov 2021 35. PRIMAX 8k. https://www.pimax.com/. Accessed 15 Nov 2021 36. Valve Index. https://store.steampowered.com/valveindex. Accessed 15 Nov 2021 37. HP Reverb G2. https://www.hp.com/us-en/vr/reverb-g2-vr-headset.html. Accessed 15 Nov 2021 38. Oculus Quest 2. https://www.oculus.com/quest-2. Accessed 15 Nov 2021 39. Varjo. https://varjo.com/. Accessed 15 Nov 2021 40. Mechatraining. https://www.youtube.com/channel/UCNvNUgGFiqf_LCCJtpfYe6g/fea tured. Accessed 15 Nov 2021 41. Microsoft HoloLens 2. https://www.youtube.com/watch?v=d3YT8j0yYl0. Accessed 15 Nov 2021 42. Microsoft Mesh. https://www.youtube.com/watch?v=Jd2GK0qDtRg. Accessed 15 Nov 2021 43. Mercedez Benz. https://www.youtube.com/watch?v=sUPgT_jGsQw. Accessed 15 Nov 2021 44. Automotive. https://www.youtube.com/watch?v=mWaQfjEJIMQ. Accessed 15 Nov 2021 45. PTC Vuforia. https://www.ptc.com/en/blogs/corporate/augmented-reality-strategy. Accessed 15 Nov 2021 46. Matterport. https://matterport.com/. Accessed 15 Nov 2021 47. 3Data. https://3data.io/platform/. Accessed 15 Nov 2021 48. Virtualics. https://virtualitics.com/products/immersive-platform/. Accessed 15 Nov 2021 49. Bublar Group. https://www.auganix.org/bublar-group-partners-with-dafo-to-create-virtualreality-fire-emergency-training-solution/. Accessed 15 Nov 2021 50. Kit-AR. https://kit-ar.com/. Accessed 15 Nov 2021 51. Glartek. https://glartek.com/. Accessed 15 Nov 2021 52. Immerse. https://immerse.io/. Accessed 15 Nov 2021 53. Re’flekt. https://www.re-flekt.com/. Accessed 15 Nov 2021 54. Pixo VR. https://pixovr.com/. Accessed 15 Nov 2021 55. Glue. https://www.glue.work/. Accessed 15 Nov 2021 56. Spatial. https://spatial.io/. Accessed 15 Nov 2021 57. Microsoft Mesh. https://www.microsoft.com/en-us/mesh. Accessed 15 Nov 2021 58. Arthur, https://www.arthur.digital/. Accessed 15 Nov 2021 59. SightCall. https://sightcall.com/. Accessed 15 Nov 2021 60. LibreStream. https://librestream.com/. Accessed 15 Nov 2021 61. TeamViewer. https://www.teamviewer.com/en/. Accessed 15 Nov 2021 62. 8th Wall. https://www.8thwall.com/. Accessed 15 Nov 2021 63. BlippAR. https://www.blippar.com/. Accessed 15 Nov 2021 64. The Wild. https://www.thewild.com/. Accessed 15 Nov 2021 65. IrisVR. https://irisvr.com/. Accessed 15 Nov 2021 66. Unity Reflect Review. https://unity.com/products/unity-reflect-review. Accessed 15 Nov 2021 67. Mira Prism Pro. https://mirareality.com/. Accessed 15 Nov 2021 68. Lenovo ThinkReality. https://techtoday.lenovo.com/us/en/solutions/thinkreality. Accessed 15 Nov 2021

How Virtual and Augmented Reality Can Boost Manufacturing

37

69. Epson Moverio. https://epson.com/moverio-augmented-reality. Accessed 15 Nov 2021 70. Vuzix Blade. https://www.vuzix.com/products/vuzix-blade-smart-glasses-upgraded. Accessed 15 Nov 2021 71. RealWear Navigator 500. https://www.realwear.com/navigator/. Accessed 15 Nov 2021 72. Magic Leap. https://www.magicleap.com. Accessed 15 Nov 2021 73. N Real. https://www.nreal.ai/. Accessed 15 Nov 2021 74. Nike Air. https://www.youtube.com/watch?v=0TcCqnFMaAU. Accessed 15 Nov 2021 75. Burger King – Burn that ad. https://www.youtube.com/watch?v=Hh-cx-cCO1U. Accessed 15 Nov 2021 76. IKEA – AR Furniture. https://www.youtube.com/watch?v=UudV1VdFtuQ. Accessed 15 Nov 2021 77. Google – Maps. https://youtu.be/QW1QT7DOOdA. Accessed 15 Nov 2021 78. Virtual Reality Motor Maintenance by Mechatraining LLC. https://www.youtube.com/watch? v=dq2RSlslQcU. Accessed 15 Nov 2021 79. National Geographic – AR. https://youtu.be/5QDB7CDD5aA. Accessed 15 Nov 2021 80. National Geographic – VR (360°). https://youtu.be/sPyAQQklc1s. Accessed 15 Nov 2021

Advanced Manufacturing Technologies

CAD/CAM Process Chain for Hybrid Additive Manufacturing Fabian Soffel1(B) , Sergei Egorov2 , Dominik Keller2 , and Konrad Wegener2 1 inspire AG, Technoparkstrasse 1, 8005 Zurich, Switzerland

[email protected] 2 Institute of Machine Tools and Manufacturing, ETH Zurich, Leonhardstrasse 21, 8092 Zurich,

Switzerland

Abstract. Hybrid additive manufacturing (AM) by direct metal deposition (DMD) and milling combines the advantages of both additive and subtractive processes for part fabrication. While the additive process ensures a high level of design flexibility and material efficiency, the consecutive milling steps realize the final surface quality. However, the great engineering effort for hybrid AM processes currently limits their application in several industries. Time-consuming trial and error approaches need to be replaced by model-based manufacturing strategies throughout the entire production process. The present study focuses on a CAD/CAM process chain that includes machine characterization techniques and control of the heat input. Within the machine characterization, the alignment of the melt pool to the milling tool is investigated. To control the heat input of the AM process, the effectiveness of a geometry-based approach is tested. Finally, the entire process chain is evaluated by the fabrication of a demonstrator part. The results show that the integration of machine characterization and heat input control increases the robustness of the hybrid AM process. The proposed process chain may therefore lead to a reduction of engineering effort for industrial applications. Keywords: Additive manufacturing · Direct metal deposition · Milling · Computer-aided manufacturing

1 Introduction Direct metal deposition (DMD) is an AM process where a laser creates a melt pool into that metallic powder particles are blown through a nozzle to generate welding tracks, layers and structures. As-deposited material typically has a rough surface that according to DebRoy et al. [1] needs to be post processed for high end applications. Hybrid AM machines that include both subtractive machining and AM processing equipment can substantially reduce the lead time for part generation and post processing by avoiding the transfer from one machine type to another [2]. As shown by Rettberg and Kraenzer [3], these machine configurations also allow the segment-wise fabrication of complex-shaped parts and can overcome accessibility issues of conventional manufacturing techniques. However, the control of the DMD process for the generation of arbitrary part geometries poses various challenges. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 41–46, 2023. https://doi.org/10.1007/978-3-031-33890-8_3

42

F. Soffel et al.

One challenge is the determination of processing parameters to create structurally sound parts for a specific application. As of today, time-consuming trial and error procedures are common in metal AM process development. As described by Wei et al. [4], digital twins of the AM process may reduce the engineering effort by shrinking the parameter space and shortening the overall lead-time for parameter determination. DebRoy et al. [5] indicate that these digital twins need to include various sub models, for example to predict the temperature profiles during the buildup process. One important criterion for application of those sub models is their computational cost and in many cases simplified models need to be implemented. Eisenbarth et al. [6] demonstrated how a simplified geometry-based heat input control approach can be used to fabricate parts with varying cross section areas by DMD. They successfully validated their approach by 3D scanning of the built parts and comparison of the result with the original CAD model. A previous study of the authors [7] used the same heat input control strategy to generate vertical cylinders, from which tensile test specimens were machined. It was shown that the tensile strength of both DMD and hybrid AM material exceeded that of conventionally cast parts, which confirmed the applicability of their heat model for massive parts. However, both last-mentioned studies did not apply the heat input control strategy for hybrid AM part geometries with significantly different wall thicknesses within one layer. The aim of the present study is to integrate machine characterization and simplified thermal modelling approaches to a hybrid AM process chain. The process chain is applied to a demonstrator part whose wall thickness varies between 1.1 and 15.0 mm to exploit the limits of the heat input control strategy. The results show that the active laser power adaption by the above mentioned control strategy leads to a robust DMD process and defect-free surfaces after milling in both the massive and the thin-walled sections. Therefore, the proposed process chain proves to increase the robustness of the hybrid AM process and may reduce the engineering effort for future industrial applications.

2 Materials and Methods 2.1 CAD/CAM Process Chain Several sub-steps are included in the proposed CAD/CAM process chain as illustrated in Fig. 1. Within the design workspace of the CAD software, the final part geometry and an oversized geometry as stock model are created (Fig. 1a). On the hybrid AM machine tool the offset between the laser and milling centreline is determined by milling a ring and welding a single spot at the same coordinates. The misalignment is then measured by optical microscopy and image analysis (Fig. 1b). During the part geometry import (Fig. 1c), the offset is compensated by shifting the part in the x-y-plane. To create the DMD tool path, the model is sliced into layers of pre-set height and the raster and contour laser paths are calculated for each layer (Fig. 1d). The contour paths are located directly on the part contour to obtain additional material of theoretically half the melt pool width for the subsequent milling operations. During the discretization step, the tool path is converted to a point cloud (Fig. 1e). At these points, a heat model calculates the required heat input (Fig. 1f), with which the numerical control (NC) code for the DMD process

CAD/CAM Process Chain for Hybrid Additive Manufacturing

43

is created with variable laser power. For the cutting operations, top and contour milling steps are applied within the milling CAM software (Fig. 1g, Fig. 1h).

Fig. 1. Overview of the proposed CAD/CAM process chain: (a) CAD modeling of the part and stock model, (b) determination of the melt pool offset, (c) STL import of the part geometry, (d) DMD tool path calculation, (e) tool path and material discretization, (f) heat modelling to adjust the laser power, (g) top and (h) contour milling tool path calculation.

2.2 Experimental Validation The experiments were carried out on a 5-axis machining centre Mikron HPM450U by GF Machining Solutions. The machining centre was equipped with an IPG Photonics 1 kW YLR-1000 fibre laser system and a Hybrid Manufacturing Technologies (HMT) processing head that can deposit welding tracks with a width of approximately 2.2 mm. Mild steel S235JRC (1.0122) plates were used as substrate and gas-atomized Inconel 718 powder from Oerlikon Metco as deposit material. The part model and milling tool paths were created in Autodesk Fusion 360. As a test part geometry, a 15 mm high and 15 mm wide massive cylinder with two 1.1 mm thin and 15 mm long walls was extruded based on a 2D profile. For the offset determination, a NX-NVDS 6 mm milling tool with penetration edge by Fraisa was plunged 0.4 mm into the substrate and at the same coordinates, a single welding spot was created with a laser power of 1000 W for a duration of 0.5 s. For the DMD tool path calculation, discretization and heat modelling, a self-developed research CAM (RCAM) software was used as described by Eisenbarth et al. [6, 8, 9]. A contour feed rate of 200 mm/min, raster feed rate of 335 mm/min, and layer height of 0.8 mm were selected as constant cladding parameters, and the laser power was varied between 1000 W and 490 W for a demonstrator part with variable laser power. Additionally, a reference part was fabricated with a constant laser power of 1000 W. The height profiles of the DMD parts were measured with a laser line sensor from Micro-Epsilon (scanCONTROL 2900-50/BL). Argon was used as a powder carrier, nozzle protection and shielding gas during the deposition process. Top milling

44

F. Soffel et al.

was performed with the circumference cutting edges of a 10 mm end mill by tilting the machine tool table by 90° and contour milling with a 6 mm cylindrical end mill for the upper 10 mm of the test part with the table in neutral position.

3 Results and Discussion The misalignment of the milling and the laser centreline was measured to be 196 µm in positive x direction and 153 µm in negative y direction of the machine tool. This offset was compensated during the import of the STL part geometry within the RCAM software. After tool path calculation and discretization, a geometric factor was calculated for each point as a measure for the local part massiveness. Figure 2a shows the result of this calculation. While the lowest layers are considered as fully massive due to the underlying substrate plate, there is a large variation of the geometric factor in the upper layers. Figure 2b illustrates the computed laser power profile derived from the geometric factor. In the upper regions, the laser power is more than 400 W lower at the part corners compared to the part region towards the built plate and the center of the cylinder.

Fig. 2. (a) Calculation of the geometric factor as a measure for the conductive heat flow and (b) laser power profile for part fabrication by DMD.

Figure 3a shows the test part fabricated with variable laser power and Fig. 3b the reference part with constant laser power in as-built condition after the DMD process. A diagonal line on the part surface indicates the start and end points of the contour paths, which were shifted for each layer to avoid instabilities. The height profiles of both parts obtained by laser scanning are depicted in Fig. 3c. With constant laser power, part overheating during the deposition process resulted in larger melt pool dimensions and longer solidification times. In the massive sections, this led to an increased powder efficiency and increased structure height. In the thin-walled sections, the larger melt pools and longer solidification times caused the liquid material to flow sideways, resulting in smaller structure heights and an instable deposition process, with which the minimum part height as indicated in Fig. 3c could not be achieved. With variable laser power, the part height ranges from 15.2 mm to 16.3 mm and is within the requirements for the subsequent milling operations. Therefore, the model-based heat input control strategy significantly stabilized the deposition process. It is assumed that the combination

CAD/CAM Process Chain for Hybrid Additive Manufacturing

45

Fig. 3. (a) Side view of the part with variable laser power, (b) the reference part with constant laser power, and (c) the measured height profiles.

of model-based feed-forward control with sensor-based closed-loop laser power control as presented for example by Rodríguez-Araújo et al. [10] can increase the process robustness even further for future applications.

Fig. 4. Final milled part with reference parts in as-built condition after the DMD process.

46

F. Soffel et al.

The final demonstrator part after milling is shown in Fig. 4 together with the two previously described specimens in as-built condition. The milling operations lead to very smooth and uniform surfaces. No defects were revealed from the previous deposition process. The material removal during contour milling was uniform as can be observed from the DMD material left at the bottom of the demonstrator part.

4 Conclusions The proposed process chain for hybrid additive manufacturing was demonstrated successfully. The results showed the necessity of heat input control strategies for the fabrication of parts with varying wall thickness and proved that a geometry-based approach increases the robustness of the deposition process. Therefore, the proposed process chain may lead to reduced engineering effort and increased process efficiency for industrial applications. Acknowledgments. The authors would like to acknowledge the contribution of the funding agency Innosuisse (grant number 50844) and of the companies Fraisa SA and ABB Schweiz AG, Turbocharging.

References 1. DebRoy, T., et al.: Additive manufacturing of metallic components – process, structure and properties. Prog. Mater. Sci 92, 112–224 (2018) 2. Flynn, J.M., Shokrani, A., Newman, S.T., Dhokia, V.: Hybrid additive and subtractive machine tools – research and industrial developments. Int. J. Mach. Tools Manuf. 101, 79–101 (2016) 3. Rettberg, R., Kraenzler, T.: Hybrid manufacturing: a new additive manufacturing approach for closed pump impellers. In: Industrializing Additive Manufacturing, pp. 146–159 (2021) 4. Wei, H.L., et al.: Mechanistic models for additive manufacturing of metallic components. Prog. Mater. Sci. 116 (2020) 5. DebRoy, T., Zhang, W., Turner, J., Babu, S.S.: Building digital twins of 3D printing machines. Scr. Mater. 135, 119–124 (2017) 6. Eisenbarth, D., Soffel, F., Wegener, K.: Geometry-based process adaption to fabricate parts with varying wall thickness by direct metal deposition. In: Proceedings of the 1st International Conference on Progress in Digital and Physical Manufacturing (ProDMP 2019), pp. 125–130 (2019) 7. Soffel, F., Eisenbarth, D., Hosseini, E., Wegener, K.: Interface strength and mechanical properties of Inconel 718 processed sequentially by casting, milling, and direct metal deposition. J. Mater. Process. Technol. 291, 117021 (2021) 8. Eisenbarth, D., Wirth, F., Spieldiener, K., Wegener, K.: Enhanced toolpath generation for direct metal deposition by using distinctive CAD data. In: Meboldt, M., Klahn, C. (eds.) AMPA 2017, pp. 152–161. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-66866-6_15 9. Eisenbarth, D.: Buildup strategies for additive manufacturing by direct metal deposition. Dissertation, ETH Zurich (2020) 10. Rodríguez-Araújo, J., Garcia-Diaz, A., Panadeiro, V., Knaak, C.: Uncooled MWIR PbSe technology outperforms CMOS in RT closed-loop control and monitoring of laser processing. In: Proceedings of Imaging and Applied Optics 2017, ATh2A.2 (2017)

On the Quality of Electron Beam Melted Thin-Walled Parts with Curved Surfaces Gabriele Piscopo(B) , Alessandro Salmi, Eleonora Atzeni, Luca Iuliano, Giovanni Marchiandi, Adriano Nicola Pilagatti, Giuseppe Vecchi, and Mirna Poggi Department of Management and Production Engineering, Politecnico di Torino, Turin, Italy [email protected]

Abstract. Electron Beam Powder Bed Fusion (EB-PBF) is an additive manufacturing process that uses an electron beam to melt a thin layer of metal powder and it is used to manufacture high-melting temperature alloys such as titanium alloys. Parts produced by EB-PBF suffer of low resolution and high surface roughness. In this work, the effect of geometrical parameters on the quality of unsupported overhangs was analyzed. Outcomes showed that the concavity and the surface tangent angle significantly influence the quality of the overhang. Moreover, as the tangent angle decreased further geometrical constraints were required in order to improve the surface quality. The results of this research can be used as guidelines during the design phases in order to obtain components characterized by a good surface quality. Keywords: Additive Manufacturing · Electron Beam Melting · Surface quality · Titanium alloy · Unsupported overhangs

1 Introduction Electron Beam Powder Bed Fusion (EB-PBF) also known as Electron Beam Melting (EBM) [1] is an Additive Manufacturing (AM) process in which a focused electron beam is used to melt a thin layer of metal powder in order to produce components layer-by-layer [2] of high-melting temperature alloys and it is commonly used for the production of titanium components for biomedical and aerospace applications. EB-PBF process allows obtaining complex geometries with minimal support structures volume, that in most of the cases, are not necessary [3]. In fact, in EB-PBF, the powder bed is maintained at high temperatures and as a consequence, the residual stresses that caused warping are very low. Thanks to its powerful characteristics, EB-PBF process is used for the production of final parts [4]. However, one of the main problems is related to the poor surface quality of the part that could require additional finishing operations [5, 6]. Consequently, a material allowance has to be added for next finishing or to fine tune possible distortions that occur during the process and are not compensated in the design of the part [7, 8]. From the literature review, it emerges that most of the studies were focused on mechanical and microstructural characterization and the investigation of strategies for © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 47–53, 2023. https://doi.org/10.1007/978-3-031-33890-8_4

48

G. Piscopp et al.

the minimization of surface roughness [9–11]. Only a few studies concerned the characterization of the surface quality and moreover, most of them were focused on the minimization of surface roughness by varying the process parameters [12, 13] or the geometrical variables [14, 15]. Furthermore, studies of thin walls are limited to vertical flat surfaces. In this work, the effect of geometrical parameters on the quality of curved thin wall overhangs produced by EB-PBF process in Ti6Al4V was evaluated. The overhangs were produced for concave and convex surfaces by varying geometrical parameters according to a 33 full factorial Design of Experiments (DoE) plan. After production, samples were inspected using an optical measurement system and the surface quality was evaluated using an index previously developed by the authors.

2 Material and Methods In the following sections, the overhangs characteristics, the equipment and the methodology used in the experimental investigation are described. 2.1 Overhang Characterization and Production The overhang geometries represented in Fig. 1 were defined as a function of the wall thickness (t), the surface radius (R), the minimum tangent angle (α) and the concavity/convexity of the surface. The values of the selected variables, reported in Table 1, were choosen according to the common values used in EB-PBF for the production of thin-walled geometry. After the CAD modelling, samples were produced by means of an Arcam EBM A2X system, using the standard process parameters [7]. Table 1. Geometrical factors and relative values used in DoE. Factors

Levels

Thickness, t (mm)

1–2–3

Surface radius, R (mm)

4–4.5–5

Tangent angle, α (°)

15–22.5–30

Curvature

Concave–Convex

2.2 Analysis of Thin-Wall Curved Surfaces The downward-facing surfaces of each component were measured using an ATOS Compact Scan (GOM, Braunschweig, Germany). The accuracy of the scanner is of about 0.02 mm [16]. Each sample was covered with a thin layer of talc in order to reduce the reflectivity of the metal material and to facilitate the acquisition of the point cloud. After eliminating erroneous data, the acquired point cloud was used to mathematically describe the actual geometry of the surface and to perform the deviation analysis. The

On the Quality of Electron Beam Melted Thin-Walled Parts

49

Fig. 1. Geometrical parameters of the thin-walled part (a), which dimensions are symmetrical about the centerline, and arrangement of the samples on the start plate (b).

deviation analysis, obtained by comparing the actual surface with the nominal one, was performed using the software GOM Inspect Suite 2020. To avoid an undesired edge effect, the deviation analysis was performed on a restricted area excluding 0.5 mm from the border of the nominal CAD surface, by comparing the actual geometry with the nominal one. The comparison was carried out by aligning the actual geometry with respect to the nominal one by using a best-fit procedure with the purpose of minimizing the overall deviation. The outcome of the comparison between the two geometries gives important information relating the dimensional deviation such as the maximum distance (MAX), minimum distance (MIN), mean distance, distance standard deviation, area of valid distance (AVD), integrated absolute distance, and the integrated distance (ID), the last that describes the volumetric deviation of the surface. The results of the deviation analysis were then used to describe the quality of the analyzed surface by means of a quality index, named k. The k-index, defined by Eq. 1, was previously developed by Piscopo et al. [17] and it combines the range of variation of the deviation (r) and the mean deviation (m), as follows   (1) k = 1 + m2 × r The scanning data were then elaborated in order to compute the quality index k. The k-index was used as a response in the ANOVA to study the effect and the significance of each geometrical factors.

3 Results and Discussion The downward-facing unsupported overhangs were successfully produced, and no macroscopic defects were observed in the analyzed surfaces. After the production, samples were cleaned by means of a sandblasting process using the same power used during the production and with an air pressure of 4 bar. Due to the geometrical complexity of the analyzed samples, 15 different acquisitions were necessary to completely acquire the morphology of the geometry. Thereafter, acquisition data were elaborated in order to compute the k-index. The result of ANOVA,

50

G. Piscopp et al.

after eliminating the outliers, are reported in Table 2. Results indicated that about 88% of the variation of the selected response was explained by the present DoE. According to the statistical analysis, the most significant parameters, characterized by a P-Value lower than 0.05, are the tangent angle, the curvature, their interaction and the thickness. Since the minimum electron beam diameter is of about 0.25 mm, deviation between the nominal geometry and the actual one lower than 0.25 mm was attributed to the contour scan strategy and not to the geometrical effect. This deviation corresponds to a k-value of about 0.4, consequently, a k-value lower than 0.4 was selected as the index value for good quality. It is worth to note that the median of the k-index is around 0.44. Table 2. Results of analysis of variance for the downward-facing surfaces.

From Fig. 2 and Fig. 3 which illustrate the main effect plot and the interactions between the analyzed variables, it is possible to estimate the combination of variables in order to obtain the best quality that is concave surface, high tangent angle and large thickness. Moreover, according to our statistical analysis, it is observed that in general lower k-index values are obtained when the concave surface is used, regardless of the combinations of the other variables used. In detail, analyzing the effect of tangent angle, it is possible to point out that using a value of α = 30° the value of k-index is lower than 0.4, which corresponds to good surface quality, for each combination of the other geometric variables. Moving from α = 30° to α = 22.5°, which means have a more inclined surface, from the graph it is observed that a k-index value lower than 0.4 is obtained only for the thicker geometries. Afterwards, reducing again the tangent angle from α = 22.5° to α = 15°, it is observed that none of the geometrical variable combinations allows obtaining a k-index value lower than 0.4.

On the Quality of Electron Beam Melted Thin-Walled Parts

51

Consequently, the value α = 15° should be avoided when possible. However, if it is not possible to change the tangent angle, the use of thick walls combined with small radius should be preferred in order to obtain a better surface quality.

Fig. 2. Main effect plot for the quality index k.

Fig. 3. Interaction plot for the quality index k.

4 Conclusions This work investigates the effect of geometrical parameters on the quality of unsupported curved thin-walled parts produced by EB-PBF process. A specific index previously developed for laser-based powder bed fusion (LB-PBF) parts was used to describe the surface quality. The main results can be summarized as follows: – the k-index is helpful to evaluate the surface quality of curved thin-walled parts produced by EB-PBF process; – when the value of k-index is lower than 0.4, the dimensional deviation of the parts is comparable with the electron beam diameter; – a k-value lower than 0.4 is obtained for concave surfaces, and the optimum is observed at higher tangent angle and larger thickness;

52

G. Piscopp et al.

– for both concave and convex surfaces, reducing the tangent angle from 30° to 15°, the number of fixed parameters values necessary to obtain a good surface quality increases progressively. In fact, when α = 30° all the values of radius and thickness analyzed in this work give a good surface quality. When α = 22.5°, it is suggested to use a wall thickness of 3 mm. Finally, when α = 15°, a thickness of 3 mm combined a radius of 4 mm allows obtaining the lower value of k-index.

References 1. Kumar, S.: Electron Beam Powder Bed Fusion. In: Additive Manufacturing Processes, pp. 65– 78. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-45089-2_4 2. Galati, M., Iuliano, L.: A literature review of powder-based electron beam melting focusing on numerical simulations. Addit. Manuf. 19, 1–20 (2018) 3. Ameen, W., Al-Ahmari, A., Mohammed, M.K.: Self-supporting overhang structures produced by additive manufacturing through electron beam melting. Int. J. Adv. Manuf. Technol. 104(5– 8), 2215–2232 (2019). https://doi.org/10.1007/s00170-019-04007-3 4. Baudana, G., et al.: Titanium aluminides for aerospace and automotive applications processed by electron beam melting: contribution of politecnico di torino. Met. Powder Rep. 71(3), 193–199 (2016) 5. Dolimont, A., Rivière-Lorphèvre, E., Ducobu, F., Backaert, S.: Impact of chemical polishing on surface roughness and dimensional quality of electron beam melting process (EBM) parts. In: AIP Conference Proceedings 2018, vol. 1, p. 140007. AIP Publishing LLC () 6. Ahmed, N., et al.: Electron beam melting of titanium alloy and surface finish improvement through rotary ultrasonic machining. Int. J. Adv. Manuf. Technol. 92(9–12), 3349–3361 (2017). https://doi.org/10.1007/s00170-017-0365-3 7. Atzeni, E., Rubino, G., Salmi, A., Trovalusci, F.: Abrasive fluidized bed finishing to improve the fatigue behaviour of Ti6Al4V parts fabricated by electron beam melting. Int. J. Adv. Manuf. Technol. 110(1–2), 557–567 (2020). https://doi.org/10.1007/s00170-020-05814-9 8. Minetola, P., Stiuso, V., Calignano, F., Galati, M., Khandpur, M.S., Fontana, L.: Experimental validation of laser powder bed fusion simulation. IOP Conf. Ser. Mater. Sci. Eng. 1091(1) (2021) 9. Galati, M., Minetola, P., Rizza, G.: Surface roughness characterisation and analysis of the Electron Beam Melting (EBM) process. Materials 12(13), 2211 (2019) 10. Karlsson, J., Snis, A., Engqvist, H., Lausmaa, J.: Characterization and comparison of materials produced by Electron Beam Melting (EBM) of two different Ti–6Al–4V powder fractions. J. Mater. Process. Technol. 213(12), 2109–2118 (2013) 11. Murr, L.E., et al.: Characterization of titanium aluminide alloy components fabricated by additive manufacturing using electron beam melting. Acta Mater. 58(5), 1887–1894 (2010) 12. Klingvall Ek, R., Rännar, L.-E., Bäckstöm, M., Carlsson, P.: The effect of EBM process parameters upon surface roughness. Rapid Prototyping J. 22(3), 495–503 (2016) 13. Wang, P., Sin, W.J., Nai, M.L.S., Wei, J.: Effects of processing parameters on surface roughness of additive manufactured Ti-6Al-4V via electron beam melting. Materials 10(10) (2017) 14. Gruber, S., et al.: Comparison of dimensional accuracy and tolerances of powder bed based and nozzle based additive manufacturing processes. J. Laser Appl. 32(3), 032016 (2020) 15. Smith, C.J., et al.: Dimensional accuracy of Electron Beam Melting (EBM) additive manufacture with regard to weight optimized truss structures. J. Mater. Process. Technol. 229, 128–138 (2016)

On the Quality of Electron Beam Melted Thin-Walled Parts

53

16. Hosseininaveh Ahmadabadian, A., Karami, A., Yazdan, R.: An automatic 3D reconstruction system for texture-less objects. Robot. Auton. Syst. 117, 29–39 (2019) 17. Piscopo, G., Salmi, A., Atzeni, E.: On the quality of unsupported overhangs produced by laser powder bed fusion. Int. J. Manuf. Res. 15(2) (2020)

Computational Origami Based Design in 4D Printing Mohamed H. Hassan1(B) , Jatin Sharma1 , and Paulo Bartolo1,2 1 University of Manchester, Manchester, UK [email protected] 2 Singapore Centre for 3D Printing, Nanyang Technological University, Singapore, Singapore

Abstract. In 4D printing, complex active structures are produced and investigated using several materials. However, research on designing cross folding origami structures and their use for direct 3D printing with direct CAD design was not previously reported. This research focuses on designing cross folding origami structures using computational origami for 4D printing. A comparison between computational origami and CAD software designed structures is provided and discussed in terms of printability, response/recovery time, and durability. Computational origami shows a potential for future usage in the field of 4D printing with minor constraints being considered while producing the design. Keywords: Origami · 4D printing · Computational

1 Introduction Smart materials such as shape memory polymers and alloys are a special class of materials with the ability to respond to a change in the environment by changing their properties [1]. Environmental stimuli include changes in temperature, pH, pressure, and moisture [2, 3]. The combined use of 3D printing and smart materials, known as 4D printing, is an emerging domain for the fabrication of self-cleaning, self-assembly, and shock absorbers, and intelligent parts for robotics, construction, bioengineering, and aerospace applications [2, 4, 5]. Origami is the Japanese name for the art of paper folding into 3D shapes representing birds, insects, animals, humans, and abstract shapes [6]. Origami has a range of forms starting from folding a single uncut square of paper to more complex shapes such as dinosaurs [7], insects [8] and other more complex shapes. Computational origami is a subsection of science of shapes, where computational and mathematical features of origami are used to create the design [9]. In this paper we are proposing an Origami CAD software to build and visualize origami constructs through applying creases. Then a thickness is added to the built origami model giving creases less thickness than the rest of the model to allow folding when a stimulus is applied.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 54–59, 2023. https://doi.org/10.1007/978-3-031-33890-8_5

Computational Origami Based Design in 4D Printing

55

2 Origami CAD Software The Origami CAD software was developed aiming to bridge the gap between the current computational origami and 3D/4D printing, by allowing the design and integration of origami folding patterns, thus enabling to change the structure from 2D to 3D allowing for direct printing in an automatic way. The software was developed using a programming language, Python. The software achieves its target by making the crease line (folding edges) thinner than the other parts of the model allowing for folding. The user can use specific folds using an integrated library integrated in the software. Figure 1 presents the logical pathway followed by the software starting from model creation till producing and exporting the STL file.

Fig. 1. Logical pathway followed by the software

The design concepts included in the software library are based on different textile designs (Fig. 2), such as folded motif, skew and accordion pleats [10]. The chosen concept, or concepts can be directly imported from the library as a sketch to the software, where basic sketching operations such as trimming, drawing, and patterning can be used. The sketch to model process is achieved by extrusion, where thickness is added to the sketch. The crease lines designed in the sketch are extruded as a semi-circular cylinder with less thickness compared to the rest of the model, to ensure folding of the model in

56

M. H. Hassan et al.

the required directions. Finally, the model can be exported as an STL format to be 3D printed.

Fig. 2. (a–c) Origami based design concepts, (d–f) paper-based physical mode

3 Materials and Methodology 3.1 Materials and Software PETG with molecular weight of 300 g/mol and printing temperature ranging between 195 °C and 220 °C was purchased from RS components (UK). Microsoft visual studio (Microsoft, Washington, USA) was used to build the initial graphical user interface (GUI) for the Origami CAD software. 3.2 Fabrication A PETG T-model structure (Fig. 3), was designed using the proposed software and produced using a filament-based extrusion 3D printer (Adventurer 3, China) with 100% infill. Considered processing conditions are presented in Table 1.

Computational Origami Based Design in 4D Printing

57

Fig. 3. T-model structure dimensions

Table 1. Processing Conditions Parameters

Values

Layer thickness (mm)

0.15

Filament diameter (mm)

0.40

Nozzle temperature (°C)

210

Bed temperature (°C)

80

Printing speed (mm/s)

60

3.3 Shape Recovery Test for the Origami Structure Shape recovery was evaluated by heating the printed structure above the glass transition temperature (Tg), applying force to deform the structure, cooling down the structure below Tg fixing its temporary shape, removing the force, and finally heating up the structure over Tg allowing the structure to regain its permanent shape [11]. The process is split into two stages - programming and recovery. The programming stage occurs when the structures is deformed and fixed to a temporary shape, and the recovery stages occurs when the structure is reheated to regain its original shape. PETG has a glass transition temperature of 75 °C [12], and both programming and recovery were carried out at 80 °C.

4 Results and Discussion Figure 4 shows the shape recovery of a PETG printed structure in response to heat changes. Figure 4(a) shows the original printed structure while Fig. 4(b) shows the deformed shape. Figure 4(c) shows the recovered model. The recovery time for the model was measured and recorded as 10 s. Figure 5 show a skew model that was designed using the software showing the possibility of creating such patterns easily and quickly with different directions of folding. This structure could then be integrated in designs for various structures for different applications.

58

M. H. Hassan et al.

Fig. 4. Origami printed T-Model, (a) non-deformed initial model, (b) deformed model, (c) recovered model

Fig. 5. Skew design pattern, (a) non-deformed initial model, (b) deformed model, (c) recovered model

5 Conclusion This paper proposes a software that uses computational origami methods to create infinite net models with a variety of designs due to its ability to customize geometry and creases. The obtained) designs could be used to solve several problems encountered when following normal design procedures such as reduced efficiency. The software helps in reducing the complexity of the design process help increasing creativity and innovation. The software also provides a platform to validate created designs and exported directly as an STL format for 3D printing.

References 1. Kim, J.: Multifunctional smart biopolymer composites as actuators. In: Biopolymer Composites in Electronics, pp. 311–331. Elsevier, Amsterdam (2017) 2. Zhang, Z., Demir, K.G., Gu, G.X.: Developments in 4D-printing: a review on current smart materials, technologies, and applications. Int. J. Smart Nano Mater. 10(3), 205–224 (2019)

Computational Origami Based Design in 4D Printing

59

3. Bashir, M., Lee, C.F., Rajendran, P.: Shape memory materials and their applications in aircraft morphing: an introspective study. J. Eng. Appl. Sci. 12, 50–56 (2017) 4. Zafar, M.Q., Zhao, H.: 4D printing: future insight in additive manufacturing. Met. Mater. Int. 26(5), 564–585 (2019). https://doi.org/10.1007/s12540-019-00441-w 5. Tamay, D.G., et al.: 3D and 4D printing of polymers for tissue engineering applications. Front. Bioeng. Biotechnol. 7, 164 (2019) 6. Kasahara, K., Takahama, T.: Origami for the connoisseur. Japan Publications (1998) 7. Pan, J.L., Bardwell, J.C.: The origami of thioredoxin-like folds. Protein Sci. 15(10), 2217– 2227 (2006) 8. Baek, S.-M., et al.: Ladybird beetle–inspired compliant origami. Sci. Robot. 5(41) (2020) 9. Demaine, E.D., Demaine, M.L.: Recent results in computational origami. In: Origami3: Third International Meeting of Origami Science, Mathematics and Education (2002) 10. Jackson, P.: Folding Techniques for Designers. Hachette UK, London (2011) 11. Heuchel, M., et al.: Relaxation based modeling of tunable shape recovery kinetics observed under isothermal conditions for amorphous shape-memory polymers. Polymer 51(26), 6212– 6218 (2010) 12. Hassan, M.H., Omar, A.M., Daskalakis, E., Liu, F., Bartolo, P.: Preliminary studies on the suitability of PETG for 4D printing applications. In: 7th International Conference of Materials and Manufacturing Engineering (ICMMEN 2020), p. 6 (2020)

Novel Extrusion Based Co-axial Printing Head for Tissue Engineering Jiong Yang1(B) , Wajira Mirihanage2 , and Paulo Bartolo1 1 Advanced Manufacturing, Department of Mechanical, Aerospace and Civil Engineering,

The University of Manchester, Manchester, UK [email protected] 2 Department of Materials, School of Natural Sciences, The University of Manchester, Manchester, UK

Abstract. Additive manufacturing (AM) is a key technology for the fabrication of tissue engineering scaffolds, 3D structures that support cell attachment, proliferation, and differentiation. However, commercially available printing heads for tissue engineering applications present limitations concerning the fabrication of complex multi-material hierarchical structures resembling natural tissues. This paper presents a novel printing head assembled with a new designed co-axial nozzle that aims to generate core-shell filaments. Compared to commercially available co-axial extruders, which usually use needle-based nozzles and can only process liquid materials, the novel co-axial printing head is able to process both lowviscosity and high-viscosity materials. Two independent extruding systems that can be replaced by screw-based or pressure-based extruders are assembled on the co-axial nozzle allowing different type of materials to be processed. An expanding chamber is placed at the cross section of the inner and outer channel to achieve a symmetrical velocity distribution. Simulation results demonstrate that the novel co-axial nozzle allows a balanced fluid flow enabling uniform co-axial filaments. Keywords: Co-axial Extrusion · Tissue Engineering · Core-shell Fibres · Multi-Material

1 Introduction Tissue engineering is a multidisciplinary field that combines engineering, biological, and material science, which has been widely researched in recent years to address the organ shortage problems [1]. The scaffold-based strategy is the most relevant tissue engineering approach and involves additive manufacturing (AM), biodegradable and biocompatible materials, growth factors and other biomolecules and cells [2]. In this approach, scaffolds, which are 3D physical supports for cells to attach, proliferate and differentiate, must fulfill several requirements/properties, including mechanical properties (sufficient stiffness and strength to withstand physiological stresses), chemical properties (biodegradable and degrading into non-toxic products), biological properties (biocompatible, able to create a proper biomechanical environment for new tissue formation by mimicking the © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 60–71, 2023. https://doi.org/10.1007/978-3-031-33890-8_6

Novel Extrusion Based Co-axial Printing Head

61

biological extracellular matrix (ECM)), and an appropriate macro and microstructure. Moreover, scaffolds must present a fully interconnected porous structure [3]. Therefore, with single material filament printing is quite often impossible to meet all these demands, and novel scaffold fabrication strategies allowing the fabrication of multi-materials and complex 3D scaffolds have been proposed. Hybrid extruding systems have been also designed to create scaffolds with a structure closer to the hierarchical structure of biological ECM. Co-axial extrusion systems to print core-shell filaments have been also developed. However, current commercial machines [4–8] are usually needle-based nozzles that can only process liquid-based materials and not high-viscosity materials often required to produce scaffolds for example for bone applications. This paper presents a novel co-axial extrusion system to print core-shell fibres. Contrary to traditional needle-based nozzle systems, a novel geometry for a co-axial nozzle (negative protrusion compression co-axial nozzle) was designed. An expanding chamber is placed at the cross section of the inner and outer channels aiming to generate a balanced fluid flow. A screw-based extruder was designed and assembled on the printing head to process high-viscosity materials and a heating system was considered for melting the materials. In addition, a pressure-based extruder is also considered to process low-viscosity materials. The co-axial nozzle and two independent extruders are symmetrically designed that they can be easily replaced with different types of extruder combinations, enabling to produce core-shell filaments with two different types of materials. Computational fluid dynamics (CFD) simulations were performed to evaluate this design. Numerical results showed the proposed system is able to generate a balanced core-shell filament.

2 Theoretical Design of Co-axial Nozzle Geometry 2.1 Negative-Protrusion Compression Co-axial Nozzle A co-axial nozzle is normally assembled with two needles with different diameters and matching as a core-shell structure. In a conventional nozzle, two materials from the inner and the outer channel are simultaneously extruded. Figure 1. Illustrates the conception of conventional co-axial extrusion and presents a commercial co-axial nozzle used for example to create core-shell filaments based on alginate (shell area) and a solution containing cells (core area). The filaments were printed in a cross-linking medium to provide the right strength to the filament. Cornock et.al [9] presented a protrusion co-axial printing head structure (see Fig. 2.) that refers the internal and external material are extruded and mixed by stage. In this geometry, the inner pipe is slightly longer than the outer pipe, and it protrudes around 300–500 µm from the exterior pipe. The authors demonstrated that the reduction of the fibre offset of the core-shell filament is achievable by adjusting the protrusion distance. Based on this conception, we are proposing in this paper a new geometry called negative compression protrusion nozzle in which the inner channel is shorter than the outer channel, allowing the co-axial nozzle the generate the core-shell structure before extrusion. Previous studies conducted by the authors demonstrated that a compression structure can is benefit to improve the anisotropy of the filament by squeezing the molecular chains, thus to control the printed filament properties. Negative-protrusion compression

62

J. Yang et al.

Fig. 1. (a) Needle-based co-axial extrusion nozzle; (b) commercial needle-based co-axial nozzle and an example of an application [4].

Fig. 2. (a) Section view and (b) 3D modelling of the all-in-one co-axial nozzle designed and developed by Cornock [9].

co-axial nozzle is firstly presented based on this considering, which the outer pipe is lengthened and then shrank at the exit, allowing generating a core-shell structure inside the nozzle before extruding and supply a compress of polymer materials. Figure 3 illustrates the conception of the compression nozzle. During the compression process, disordered material molecules will gradually be compressed to a relatively parallel state, which is used to increase the anisotropy of the filament. The compression ratio can be defined and calculated as follows: ε=

π ( OutletDiamter ) 2

2

π ( OriginalFluidDiameter ) 2

2

× 100% = (

d 2 ) × 100% DOut

(1)

Novel Extrusion Based Co-axial Printing Head

63

The compression efficiency indicates how much time the material needs to go through the compression region under the same extruding velocity and the same compression ratio. Compression efficiency is defined as the compression ratio divided by the compression length: η=

ε l

(2)

where ε is the compression ratio and l represents the compression length.

Fig. 3. Schematic view of the negative protrusion compression co-axial nozzle geometry, (a) inner shorter pipe nozzle and (b) inner longer pipe nozzle.

2.2 CFD Simulations CFD simulations were conducted to evaluate the novel co-axial system. The ANSYSFluent (ANSYS, USA) multi-phase model was used to visualize the material distribution and the core-shell shape. Only the top end of the nozzle and the outside space were considered in the simulations. Equations for conservation of mass, momentum, and temperature were solved in ANSYS. Velocity difference between the internal and external inlet is considered as the initial boundary condition to research the co-axial filament printability. Two high-viscous biopolymers (polycaprolactone, PCL, and polylactic acid, PLA) widely used for tissue engineering applications were considered for the simulations. Material properties are presented in Table 1. Small nozzle dimensions and free space were considered in the simulations, where the air pressure reaches a steady-state value

64

J. Yang et al.

and the influence of pressure drop and air compressibility can be ignored. In addition, the following assumptions were considered: • • • •

The material and air are considered incompressible; Pressure drops along the nozzle wall is ignored; There is no slip between the material and the needle wall; Temperature and energy are applied in simulation (Air-293K(20◦ C); PLA533K(260◦ C); PCL-363K(90◦ C), and room temperature-293K(20◦ C))

Table 1. Material Properties of CFD simulations Material Properties

Air (Initial Material)

PCL

PLA

Density (kg/m3 )

1.225

1145

1251

Specific heat capacity (J/kg · K)

1006.43

1450

1800

Dynamic viscosity (Pa · s)

1.7894E-5

3

5

Thermal Conductivity (W/m · K)

0.0242

0.14

0.11

Molar Mass (kg/mol)

28.966

68

72.1

Static Multi-phases Simulation. In these cases, the co-axial nozzle is placed above the platform and keeping at the same position. Simulation results showed that the negativeprotrusion compression co-axial nozzle can perfectly form a core-shell structure inside the nozzle and extrude the materials as a co-axial fibre. As observed the filament diameter expands after extruding (die-swell phenomena). A large space between the nozzle and platform was considered in order to investigate the material shape after extruding. A relatively reasonable distance (400μm) is used in dynamic printing simulations (see Fig. 5). Results presented in Fig. 4 showed that the negative protrusion compression geometry process a good extrusion state with a fine concentricity, which the internal material is basically located in the middle of the external material. Figure 4b and c indicate that the velocity variety has little effect on the extruded filament status, with a different core and shell thickness influenced by the extrusion velocity. This can be controlled by increasing or decreasing the loaded pressure and screw rotation velocity, which gives an approach to control the thickness of the internal and external material thickness of the co-axial filament. Figure 4d and e show the states of the co-axial filament with excessively different extrusion velocity. A relatively good core-shell structure can still be formed inside the compression nozzle. However, a large thickness difference between the inner and outer material occurs, causing an undesirable co-axial filament formation because of the large velocity variety. For custom designed co-axial extrusion requires a matched extrusion velocity to ensure the extruded filament is in a good core-shell structure.

Novel Extrusion Based Co-axial Printing Head

65

Fig. 4. The co-axial filament extrusion status with (a) inner velocity 100 mm/s and outer velocity 100 mm/s; (b) inner velocity 100 mm/s and outer velocity 75 mm/s; (c) inner velocity 75 mm/s and outer velocity 100 mm/s; (d) inner velocity 100 mm/s and outer velocity 50 mm/s; (e) inner velocity 50 mm/s and outer velocity 100 mm/s.

Simulation of Dynamic Printing. Dynamic mesh was used in these cases and a the coaxial nozzle was assumed to move in the horizontal direction with a printing velocity of 100 mm/s. High printing velocities result in non-uniform filaments. Figure 5e shows the result that the extrusion velocity of the internal materials is half of the external material (50 mm/s for internal material and 100 mm/s for external material). The printed filament performed an uneven structure, with intermittent core materials. In the case of using the same inlet velocity for both outer and inner materials, the coaxial filament (Fig. 5(a)) exhibits a uniform diameter and the inner material is perfectly positioned at the centre of the filament with a well-defined outer layer. As observed from the other figures (Fig. 5(b–e)) the core-shell thickness can be controlled by adjusting the material extruding speed, which can performed by pressure and/or screw rotational velocity. Moreover, the filament diameter can be also controlled by adjusting the printing velocity and other parameters such as temperature.

66

J. Yang et al.

Fig. 5. Dynamic co-axial printing results with velocity variety (a) inner velocity 100 mm/s and outer velocity 100 mm/s; (b) inner velocity 100mm/s and outer velocity 75 mm/s; (c) inner velocity 75 mm/s and outer velocity 100 mm/s; (d) inner velocity 100 mm/s and outer velocity 50 mm/s; (e) inner velocity 50 mm/s and outer velocity 100 mm/s.

3 Co-axial Nozzle Design and Optimization 3.1 Co-axial Nozzle Modelling As the new designed co-axial nozzle will be used process high-viscous materials, an allin-one nozzle is considered instead of those nozzles combining with two needle tubes. To avoid undesired solidification inside the nozzle channel, a small co-axial nozzle was considered. The co-axial nozzle is the core part of the co-axial extrusion system that connects two separated extruding chambers and matches two materials into a core-shell fibre. The nozzle structure has a relatively small dimension (less than (1 mm) at the outlet point, which requires a higher fabricating precision around (100 − 300 μm). The 3D representation of the compression co-axial nozzle is presented in Fig. 6. This initial concept was slightly modified by developing a co-axial nozzle with an expanding chamber at the cross section of the inner and outer pipe (Fig. 6b). Due to the obstruction of the internal pipe, the fluid distribution in the external pipe is uneven at the co-axial matching region. Therefore, the expanded chamber aims to allow more fluid entering the right-hand side allowing forming a uniform distribution. Moreover, the nozzle will be combined with a pressure-assisted extruder and a screw-based extruder allowing

Novel Extrusion Based Co-axial Printing Head

67

to process a wide range of low and high viscous materials. Therefore, it is important to consider a simple assembly method to install the nozzle and the pressure-based or screw-based extruders. The upper region will be used to install the extruders with a seal part. Symmetrical structure is designed at this region, hence two extruders can be installed connecting the inner or outer channel. The matching region will slowly combine two independent fluid flows into a co-axial structure. The above investigated geometry (negative protrusion compression geometry) is designed at the top end of the nozzle.

Fig.6. Section view of new designed all-in-one compression co-axial nozzle (a) normal structure co-axial nozzle; (b) with an expanding chamber co-axial nozzle.

3.2 Evaluation of CFD Simulations 3D CFD simulations are applied to evaluate the designed co-axial nozzle structure. Two high-viscosity materials (PLA and PCL) are also applied for both inner and outer channel respectively. The outlet pressure is the default air pressure (101,325 pa) and pressure-inlet model is used for 3D simulations. 6 bar (600 kPa) is used for both inlet pressure, which is normally used to operate with a bio-additive manufacturing system. The materials are assumed at high temperature inlet (533K for PLA and 363K for PCL, respectively), which performing as the melting temperature and working temperature in the extruders. Velocity. Figure 7 presents the fluid flow velocity distribution results. In the normal structure that without expanding chamber, there is an obviously imbalanced velocity distribution at the outer channel, which the material is mainly concentrated on the left side since the obstruction of the internal pipe. The uneven velocity distribution may cause a bad concentricity and a non-uniform core-shell filament. A small expanding chamber was designed at the cross section of the inner and outer pipe to let more fluid flow into the right side. Figure 7b and c shows the velocity distribution with two different size of expanding chamber. The section view at plane L shows a relatively uniform velocity distribution with modifying the dimension of the expanding chamber. With this CFD simulation results, it is safe to assume that there is a suitable dimension of the expanding

68

J. Yang et al.

chamber that will perform a symmetrical velocity distribution for different material combinations. In addition, the co-axial nozzle with expanding chamber can make the velocity distribution at the outer channel closer to the velocity curve at the inner channel when they have the same inlet velocity. The final velocity variety is caused by the difference of inlet plane diameter between the internal and external channel. Therefore, it is obviously that this structure has a great effect on the velocity distribution of two channels that can perform a uniform and balanced fluid flow before flow into the compression region. The extrusion velocity of both fluid flow will increase rapidly in the compression region, allowing the original extrusion system to work with a relatively smaller pressure and a smaller screw rotation speed and finely control the fluid flow (Fig. 8).

Fig. 7. Modelled fluid pathways and the associated fluid velocity profiles of (a) general structure co-axial nozzle; (b) with expanding chamber nozzle; (c) with modified expanding chamber nozzle.

Temperature. Figure 9 presents the temperature profile of the co-axial nozzle with two heated material extrusions. PLA and PCL are applied for the internal and external material. The initial inlet temperature (533K for PLA and 363K for PCL) is considered as the printing temperature that usually heated in the screw-based extruders. The temperature profiles show similar results for three co-axial nozzles with or without expanding chamber. The expanding chamber has little influence on the temperature profile. The temperature at outlet plane illustrates an almost same temperature of the inner and outer material. Heat exchange occurs inside the co-axial nozzle, which the lower temperature material is slightly heated while the higher temperature material exhibited a certain level of cooling. It can be seen that both materials is maintained above the melting temperature, which promise a high-viscosity material smoothly extruding through the nozzle (Fig. 10).

Novel Extrusion Based Co-axial Printing Head

69

Fig. 8. Velocity vs fluid flow inside the co-axial nozzle.

Fig. 9. Temperature profiles of the co-axial nozzle with inner PLA (533K) and outer PCL (363K) extrusion.

70

J. Yang et al.

Fig. 10. Temperature variations vs fluid flow.

4 Conclusions A completely novel co-axial nozzle with a new presented geometry, negative protrusion compression geometry, is designed and developed. The negative protrusion geometry can form the core-shell fluid flow inside the co-axial nozzle and extrude two materials performing as a well-structured co-axial filament. The compression geometry and negative protrusion geometry have benefit on processing a good concentricity of the core-shell material distribution. Static and dynamic processing CFD simulations demonstrated our presented structure is capable to produce a fine co-axial filament. Therefore, the negative protrusion compression structure is applied in the co-axial nozzle design and an expanding chamber is placed at the cross section of the inner and outer pipe, allowing a balanced fluid flow before materials being extruded into the compression region. The all-in-one nozzle has a low temperature decrease with small dimension and heat exchange promise both high-viscosity keeping above the melting temperature inside the co-axial nozzle.

References 1. Yu, Y., Zhang, Y., Martin, J.A., Ozbolat, I.T.: Evaluation of cell viability and functionality in vessel-like bioprintable cell-laden tubular channels. J. Biomech. Eng. 135(9), 091011 (2013) 2. He, J., Shao, J., Li, X., Huang, Q., Xu, T.: Bioprinting of coaxial multicellular structures for a 3D co-culture model. Bioprinting 11, e00036 (2018) 3. Liu, W., et al.: Coaxial extrusion bioprinting of 3D microfibrous constructs with cell-favorable gelatin methacryloyl microenvironments. Biofabrication 10(2), 024102 (2018) 4. Wang, X., et al.: Coaxial extrusion bioprinted shell-core hydrogel microfibers mimic glioma microenvironment and enhance the drug resistance of cancer cells. Colloids Surf., B 171, 291–299 (2018) 5. Gao, Q., He, Y., Fu, J.Z., Liu, A., Ma, L.: Coaxial nozzle-assisted 3D bioprinting with built-in microchannels for nutrients delivery. Biomaterials 61, 203–215 (2015)

Novel Extrusion Based Co-axial Printing Head

71

´ eszkowski, W., Barbetta, A.: Co-axial wet-spinning in 3D bio6. Costantini, M., Colosi, C., Swi˛ printing: state of the art and future perspective of microfluidic integration. Biofabrication 11(1), 012001 (2018) 7. Jia, W., et al.: Direct 3D bioprinting of perfusable vascular constructs using a blend bioink. Biomaterials 106, 58–68 (2016) 8. Wang, Y., Kankala, R.K., Zhu, K., Wang, S.B., Zhang, Y.S., Chen, A.Z.: Coaxial extrusion of tubular tissue constructs using a gelatin/GelMA blend bioink. ACS Biomater. Sci. Eng. 5(10), 5514–5524 (2019) 9. Cornock, R., Beirne, S., Thompson, B., Wallace, G.G.: Coaxial additive manufacture of biomaterial composite scaffolds for tissue engineering. Biofabrication 6(2), 025002 (2014)

Localization and Control of a Mobile Robot for Additive Manufacturing Abdullah Alhijaily(B)

, Zekai M. Kilic, and Paulo Bartolo

Department of Mechanical, Aerospace and Civil Engineering, University of Manchester, Manchester, UK [email protected], {Zekaimurat.kilic, paulojorge.dasilvabartolo}@manchester.ac.uk

Abstract. Additive manufacturing is a fast-growing technology and considered one of the key technological pillars of the current industrial revolution known as Industry 4.0. Current additive manufacturing systems are based on gantries and robotic arms. However, these different mechanical systems present some limitations in terms of portability, flexibility, and size of the produced objects. This paper proposes a novel approach for additive manufacturing based on the use of mobile robots. The developed approach is based on an omnidirectional mobile robot equipped with a set of mecanum wheels. This paper describes the development of such mobile robots. It also discusses the kinematic model, localization, control, and communication system. Moreover, the paper presents the Kalman filtering of the kinematic model and the sensors (inertial measurement unit, encoders, and a global sensor that is a virtual reality tracker). Preliminary results show that the developed mobile robot presents high positional accuracy making it a suitable system for additive manufacturing. Keywords: Mobile 3D Printer · Accurate Localization · Additive Manufacturing

1 Introduction Additive manufacturing (AM) represents a group of processes that produces parts layerby-layer. Currently, AM systems comprise gantry systems and robotic arms. Gantry systems are the most common mechanical systems due to their high positional accuracy and the rigidity of their structure. However, gantries suffer from limited workspace and are not easily configured for multi-axis printing. Therefore, novel systems based on robotic arms are being developed since they allow multi-axis printing and have a larger workspace [1]. Despite the advantages of robotic arms, they require complex control, advanced path planning and have inconsistent accuracy within the workspace. Mobile robots are currently being explored as a new mechanical system to handle the 3D printing tasks [2]. They are not constrained by the machine size unlike gantries and robotic arms; thus, they cover working areas larger than their sizes. Moreover, mobile robots can be easily transported to different locations as they do not require any structural setups. However, mobile robots suffer from several external disturbances such as the slipping © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 72–81, 2023. https://doi.org/10.1007/978-3-031-33890-8_7

Localization and Control of a Mobile Robot for Additive Manufacturing

73

and skidding of the wheels. Another source of error is due to the robot hardware and assembly, such as dissimilar wheel radii and misalignment of the wheels [3]. Thus, odometry, which is estimating the robot state from the wheels’ encoders, is not reliable for long motions. Moreover, omnidirectional wheels introduce vibrations into the system since there is a finite number of rollers in the wheel [4]. These problems make precise localization and control of mobile robots a challenge. To solve these major issues of mobile AM, this paper focuses on developing localization and control modules that are accurate enough for the motion of material extrusion AM. As these modules will be later used on a mobile robot for AM, the robot must be as accurate as possible. The paper briefly reviews the current attempts to achieve high accuracy for mobile robots, proposes a novel mobile robot that is capable of performing AM motions, and discusses the results of the corresponding accuracy and precision tests.

2 Related Work Achieving high accuracy is mainly associated with the type of sensors used and the fusing algorithms implemented. Eberhard and Tang [3] fused the corrected odometry and an infra-red projector and detector sensor using the Extended Kalman Filter (EKF). In order to evaluate their localization and control systems, the robot is required to move in different motion types such as lines and circles. The localization module provided millimeter-scale accuracy, but when the robot moved far from the origin the accuracy decreased to centimeter-scale. Marques et al. [5] developed a mobile 3D printer using an optical mouse sensor without any sensor fusion algorithm. The authors reported an average error of 1.23 mm. Similarly, Palacin et al. [6] tested an optical mouse for indoor mobile robots and achieved a maximum error of 0.8 mm. However, the authors found that mouse sensors are largely dependent on the height of the sensor from the ground. They reported that a 0.1 mm variation of the height causes an error of 1% in the measurements. Such dependence on height is even worse for mecanum wheels due to vertical vibrations. Tiryaki et al. [2] fused the measurements of a camera and markers system with the control of a mobile robot using EKF. The mobile robot was also equipped with an accurate robotic arm and the full system achieved a maximum error of 9.8 mm to 3D print a large structure. Moreover, Nardi et al. [7] developed an accuracy benchmarking software for simultaneous localization and mapping algorithms. They reported a best case of 20 mm trajectory error using different devices for the benchmarking software. Drawing mobile robots have similar requirements to AM mobile robots since any error in the trajectory has a direct impact on the final product. Shih and Lin [8] developed a mobile robot and, instead of a printing head, attached a pen to allow drawing. For controlling the robot, they used inverse kinematics plus proportional control approach which showed high tracking accuracy. Their system achieved around 2 mm maximum error and averaged 1.5 mm error. However, they only relied on the odometry of wheels for localization and they assumed no slippage or backlash for the wheels. Such assumptions are not practical and will result in drifts for long motions. To address these limitations, we propose in this paper a localization system that is based on fusing three sensors using Kalman filter. We used also a Virtual Reality

74

A. Alhijaily et al.

(VR) sensor as the global sensor which is a novel approach for seeking high accuracy positioning for AM. Furthermore, the mobile robot is controlled using two proportional– integral–derivative (PID) loops to allow precise path tracking. The future goal of this mobile robot is to be used as a 3D printer for material extrusion AM. However, the accuracy and precision must first be evaluated.

3 System Design The purpose of the developed mobile robot is to execute 3D printing tasks, which requires omnidirectionality and high precision. Figure 1 shows the mobile robot and its components. For omnidirectional motion, the robot is equipped with four mecanum wheels (goBilda, United States) which allow the omnidirectional movement of the platform. The mecanum wheel has a diameter of 96 mm and contains 10 small rollers spread along the circumference of the wheel and oriented by 45°. The odometry data is acquired using the absolute encoder that is mounted on each motor (ROBOTIS, South Korea). Moreover, an IMU (LP-RESEARCH, Japan) is attached to the platform to measure the angular velocities and the linear accelerations. The absolute positioning sensor used in the mobile robot is a Vive Tracker (HTC, Taiwan) which is a VR object that is used in HTC Vive systems to give the pose of the tracked object, which is rarely used for mobile robot’s localization. The Vive Tracker calculates its position and velocity based on the light data coming from two fixed light emitting stations and an internal IMU. Finally, a drawing mechanism is attached at the front of the robot to serve as a ground truth which is the actual path traveled by the robot.

Fig. 1. Developed mobile robot and main components

The drawing mechanism will be replaced later for the components needed for material extrusion AM. A Z-axis will replace it and it will be controlled using similar motors to the one used with the wheels to ensure compatibility in the software. On the Z-axis several heating and cooling components will be used. Also, an extruder will be placed near the back of the robot to feed the filament material to the hotend. However, first, the mobile robot must be accurate and precise before installing these components, which is what is discussed in this paper.

Localization and Control of a Mobile Robot for Additive Manufacturing

75

3.1 Kinematic Model The purpose of the kinematic model is to find an equation that correlates the wheels’ angular velocities and the velocities of the mobile platform. This is needed for controlling the mobile robot by giving it the overall motion in the body frame instead of the individual velocity of each motor. The H matrix which is the inverse of the Jacobian matrix of the mobile robot of interest is found to be [9]:   ⎤ ⎡ ⎡˙ ⎤ θw1 −1 1 |dx | + dy  ⎡ ⎤ ⎢ ⎥ x˙ p ⎢ θ˙w2 ⎥ ⎥ = H q˙ p = 1 ⎢ 1 1 −|dx | − |dy| ⎥⎣ y˙ p ⎦ ⎢ (1) ⎣ θ˙w3 ⎦ r ⎣ 1 1 |dx | + d y   ⎦ ˙ θ θ˙w4 −1 1 −|dx | − dy  where θ˙wi is the rotational velocity of the i th wheel, the p subscript indicates that the variable is in the robot frame, x˙ , y˙ and θ˙ are the linear and rotational velocity of the platform, dx and dy are the distance between each wheel and the center of the robot. Moreover, the velocity in the robot frame can be transformed to the world frame using the following equation ⎡

⎤ cos(θ ) − sin(θ ) 0 q˙ = ⎣ sin(θ ) cos(θ ) 0 ⎦q˙ p 0 0 1

(2)

In this case, the rotation of the mobile robot around the Z-axis is assumed to be zero, otherwise, the system would be nonlinear, and not enforcing the angle to zero leads to drift over time, which is done in the controller. 3.2 Localization Module The localization of a mobile robot is the process of finding the position of the robot within its environment. This is done by fusing the sensors’ measurements using a probabilistic localization algorithm. The probabilistic algorithm used in this research project is the Kalman Filter (KF), which is a linear Gaussian filter that works on linear systems with gaussian uncertainties. The idea behind the KF is to predict the location of the robot using a motion model and then correct it based on a measurement model. The motion model used here is based on the data acquired from the IMU, while the measurement model is based on the Vive Tracker, encoders, and the kinematic model. The general form of the motion and measurement models are respectively xk = Ak sk−1 + Bk uk + wk

(3)

zk = Ck sk + vk

(4)

where sk is the state vector, Ak is the state transition matrix, Bk is the input matrix, uk is the control vector, zk is the measurement vector, Ck is the measurement matrix and wk

76

A. Alhijaily et al.

and vk are the motion and measurement noises, the subscript k is the current step while k − 1 is the previous step. The matrices Ak and Bk can be described as follows ⎤ ⎡ ⎡ ⎤ 1 0 0 tk 0 0 0.5tk2 0 0 ⎢ 0 1 0 0 t 0 ⎥ ⎢ 0 0.5t 2 0 ⎥ k ⎥ ⎢ ⎢ ⎥ k ⎥ ⎢ ⎢ ⎥ 0 0.5tk2 ⎥ ⎢ 0 0 1 0 0 tk ⎥ ⎢ 0 (5) Ak = ⎢ ⎥ ⎥, Bk = ⎢ ⎢0 0 0 1 0 0 ⎥ ⎢ tk 0 1 ⎥ ⎥ ⎢ ⎢ ⎥ ⎣0 0 0 0 1 0 ⎦ ⎣ 0 0 ⎦ tk 0 0 tk 000 0 0 1 where tk is the time step at time k. Finally, the measurement matrix Ck is given by

I3 03×4 Ck = (6) 04×3 H where I3 is the identity matrix of size 3, 0i×j is the zero matrix of size i × j and H is from Eq. (1). The noises of the system wk and vk , are assumed to be independent Gaussian noise with covariance R and Q respectively: p(w) ∼ N (0, R)

(7)

p(v) ∼ N (0, Q)

(8)

R and Q are chosen based on the accuracy of the used sensor. Since R is related to the IMU, it has higher diagonal values compared to Q which contains the Vive Tracker and the encoders. After developing the models, the Kalman filter algorithm can be described as [10]: sk = Ak sk−1 + Bk uk

(9)

P k = Ak Pk−1 ATk + Rk

(10)

 −1 Kk = P k CkT Ck P k CkT + Qk

(11)

sk = sk + Kk (zk − Ck sk )

(12)

Pk = (I − Kk Ck ) P k

(13)

where the bar above the symbols refers to the priori value and T is the transpose of the matrix. Kk is the Kalman gain, Pk is the estimated covariance matrix. The localization module allows the position of the robot to be estimated even if one of the sensors failed or produced corrupted signals.

Localization and Control of a Mobile Robot for Additive Manufacturing

77

3.3 Control Module The mobile robot is controlled in two control loops, the outer loop is the trajectory following controller and the internal loop is the motors’ internal PID controller. The trajectory following is a linear PI controller that is manually tuned ensuring it is fast, stable, and presents minimum oscillations and overshoot. The outer loop runs at 500 Hz while the internal controller runs at 8 kHz. First, the path planning system gives the desired pose, which is compared with the estimated pose and the resulting error is sent to the trajectory following controller at each time step. Then, the controller calculates the desired wheels’ velocities based on the kinematic model. The internal loop generates the pulse-width-modulation signal to the motors based on the desired wheels velocities and the current motors encoders measurements. After that, each sensor sends its measurement to the sensor fusion module. Finally, the outer loop is closed when the sensor fusion module implements the localization method (see Sect. 3.2) and produces the estimation based on the sensors’ measurements. These steps are presented as a block diagram in Fig. 2.

Fig. 2. The block diagram of the controllers.

3.4 Communication Module Each component of the system (e.g. motors, sensors, and algorithms) is connected to other parts using Robot Operating System (ROS). ROS is an open-source middleware dedicated to robotic applications. It provides a suite of software and libraries that are essential for developing robots. One of the functionalities provided by ROS is the communication of data messages over topics to different nodes in the system. Nodes in ROS are executables that process the program and can communicate with other nodes over topics, while a topic is a bus that transfers message data between nodes. There are many topics and nodes that are developed for the mobile robot. An example of a node in the developed mobile robot is the localization_manager node. This node subscribes to all the sensors in the system and calculates the estimated pose. Then, the localization node publishes its data to a topic that is subscribed by all nodes that need the estimation to perform their tasks, such as the controllers.

78

A. Alhijaily et al.

4 Experimental Work High accuracy and precision are two important requirements for an AM system. In the context of this paper, accuracy is the agreement between the motion of the robot and the planned path, while precision is how close the current motion of the robot are to previous similar motions. Therefore, the experimental work focused on assessing the localization and control by measuring the accuracy quantitatively and the precision qualitatively. 4.1 Accuracy Assessment For the accuracy test, the robot was requested to move at a velocity of 35 mm/s in a square with 100mm sides. Since there was a drawing pen at the tip of the mobile robot, the motion was drawn on a grid paper. The drawing was considered as the ground truth and the grids on the paper along with a caliper were used to measure the errors. Table 1 lists the errors of the Kalman filtered, Vive Tracker, and odometry positions. Table 1. Accuracy analysis based on experiment #1. Units are in mm. Error

Localization Vive Tracker Odometry (KF)

Minimum

0

0.029

0

Average

0.468

0.696

3.554

2.12

6.08

Maximum 0.969

As can be seen from Table 1, the developed localization module produced better results than that of the Vive Tracker and the odometry alone. Moreover, it maintained a sub-millimeter maximum and average errors. The experiment showed that the developed mobile robot performed better than the studies reported in the literature. Sub-millimeter errors are within the acceptable results for material extrusion of plastics using large line widths. To better mimic the toolpath of 3D printers during the fabrication process, the robot was asked to perform the motion of a sliced cylindrical object. Using Cura slicer, the 60 mm diameter cylinder was sliced using both concentric and line filling, which are commonly used for 3D printing. The robot performed the motion of two layers, one has the concentric filling and the other has lines filling. The result of this experiment is shown in Fig. 3. The lines in this figure that across the circles are the motion of the robot when moving from a circle to another. Also, there are two lines that indicate the start and end of the current layer. This experiment showed similar accuracies and errors to those listed in Table 1. 4.2 Precision Assessment To test the precision of the system, the mobile robot was requested to draw a 100 mm diameter circle 50 times in the same run at a tangential velocity of 35 mm/s. Figure 4 shows the drawings performed by the robot and the data of the KF and the Vive Tracker.

Localization and Control of a Mobile Robot for Additive Manufacturing

79

Fig. 3. The trajectory of the mobile robot when performing a motion similar to the ones that appear in 3D printing.

Fig. 4. (a) 50 circles drawn on top of each other and (b) Vive tracker measurement, KF estimation, and the desired path.

Results from Fig. 4a show that the mobile robot moves consistently with high precision and accuracy. The deviation between the desired trajectory and the ground truth is unnoticeable except for the regions on the top right where there are small deviations in only two of the 50 drawn circles. The deviations are likely due to long corrupted signals from the Vive Tracker. However, from Fig. 4b it is possible to observe that sometimes

80

A. Alhijaily et al.

the Vive Tracker produces short noisy and corrupted signals which did not significantly affect the KF.

5 Conclusion and Future Work The paper discusses the main limitations of using mobile robots as 3D printers. As shown, there is a need for better mobile robots exhibiting high accuracy and precision. Thus, we presented an omnidirectional mobile robot that can achieve sub-millimeter accuracy and high precision using various sensors which are communicating via ROS. As the robot proved that its accuracy and precision are sufficient for AM, future work will be focused on developing the mobile robot more for AM. This will be done by equipping it with the required 3D printing hardware and developing various software to convert it into an AM machine. Figure 5 shows the early works of converting the mobile robot into a capable material extrusion machine.

Fig. 5. Early works of converting the mobile robot into a 3D printer.

Acknowledgement. The first author acknowledges the support received from The Saudi Arabian Cultural Bureau to conduct his PhD.

References 1. Jiang, J., Newman, S., Zhong, R.: A review of multiple degrees of freedom for additive manufacturing machines. Int. J. Comput. Integr. Manuf. 34(2), 195–211 (2020)

Localization and Control of a Mobile Robot for Additive Manufacturing

81

2. Tiryaki, M., Zhang, X., Pham, Q.: Printing-while-moving: a new paradigm for large-scale robotic 3D printing. In: 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2019) 3. Eberhard, P., Tang, Q.: Sensor data fusion for the localization and position control of one kind of omnidirectional mobile robots. In: Multibody System Dynamics, Robotics and Control, pp. 45–73 (2012). https://doi.org/10.1007/978-3-7091-1289-2_4 4. Park, Y.K., Lee, P., Choi, J.K., Byun, K.S.: Analysis of factors related to vertical vibration of continuous alternate wheels for omnidirectional mobile robots. Intel. Serv. Robot. 9(3), 207–216 (2016). https://doi.org/10.1007/s11370-016-0196-3 5. Marques, L.G., Williams, R.A., Zhou, W.: A mobile 3D printer for cooperative 3D printing. In: Solid Freeform Fabrication Symposium – An Additive Manufacturing Conference (2017) 6. Palacin, J., Valgañon, I., Pernia, R.: The optical mouse for indoor mobile robot odometry measurement. Sens. Actuators A 126(1), 141–147 (2006) 7. Nardi, L., et al.: Introducing SLAMBench, a performance and accuracy benchmarking methodology for SLAM. In: 2015 IEEE International Conference on Robotics and Automation (ICRA), pp. 5783–5790 (2015) 8. Shih, C., Lin, L.: Trajectory planning and tracking control of a differential-drive mobile robot in a picture drawing application. Robotics 6(3), 17 (2017) 9. Lynch, K.M., Park, F.C.: Modern Robotics. Cambridge University Press (2017) 10. Thrun, S.: Probabilistic robotics. Commun. ACM 45(3), 52–57 (2002)

Development of a Large Size 3D Delta Printer for Advanced Polymers D. Pereira1 , M. Leite2 , M. Ferreira3 , D. Machado4 , R. Dionísio1 and R. A. Cláudio1,2(B)

,

1 ESTSetúbal e CDP2T, Polytechnic Institute of Setúbal, Setúbal, Portugal

[email protected]

2 IDMEC, Instituto Superior Técnico, Lisbon, Portugal 3 Department of Electrical Engineering, Polytechnic Institute of Setúbal, Setúbal, Portugal 4 Universidade Nova de Lisboa, Lisbon, Portugal

Abstract. Additive Manufacturing has been extensively used in the last years for prototyping and small series production of functional parts. One of the most popular processes is material extrusion, where a material is extruded through a nozzle, being the part created layer by layer. In this work, an advanced 3D printer is presented, build by the authors, based in a Delta Robot configuration, able to produce parts with 400 mm diameter and up to 850 mm height. The chosen concept provides a very stiff structure with ease of access to all components. A high temperature water-cooled print head (that can reach 500 °C) and a closed heated chamber makes this printer able to produce large size parts with some advanced thermoplastic materials. To improve accuracy, a new sensor based in a load cell with specific signal conditioning is proposed for z-probe and to accurately determine some characteristics of the Delta mechanism. The principal features of this new sensor are compared with a previous assembled sensor (IR). A careful assembly and non-clearance mechanisms with the new sensor proposed are expected to improve print quality. Keywords: FFF · Delta Printer · Accuracy · z-probe · advanced materials

1 Introduction Additive manufacturing is one for the most fascinating technologies for the manufacturing industry with a great potential for mass customization, [1]. The process to produce a part involves much lower number of operations minimizing the process logistics. This results in much shorter lead times (typically 5 times less) and in some cases (depending on technology and part size) lower costs, [2]. There is a large range of different processes to produce 3D parts, being classified by ASTM F2792 into 7 distinct families, [1]. Among the most popular processes is material extrusion, where a material is extruded through a nozzle, being the part created layer by layer. This is one of the most attractive processes because the large diversity of printers available in the market, covering from low cost printers for domestic users, [3], up to high-end professional printers to produce © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 J. O. Correia Vasco et al. (Eds.): ProDPM 2021, STAM, pp. 82–94, 2023. https://doi.org/10.1007/978-3-031-33890-8_8

Development of a Large Size 3D Delta Printer for Advanced Polymers

83

functional parts for functional applications like certified aeronautical components, [4]. This process was patented by Stratasys Inc., who attributed the designation of fused deposition modeling (FDM) that is a trademark of Stratasys Inc.. The fused filament fabrication (FFF) is a synonym of FDM, [5]. One of the most successful projects after Stratasys Inc. Patent expiration was RepRap created by Adrian Bowyer and his team, [3, 5]. They created self-replicating open-source 3D printers, generating a vast community of “makers” producing and developing their own 3D printers with parts produced by other printers. Some of these development kits of open-source printers are: Darwin (version I) (Fig. 1), Kossel Mini (Fig. 2), Rostock and Prusa i3. While projects like Darwin (version I) or Prusa i3 are based in a three-axis configuration, both Kossel Mini and Rostock are based in a Delta Robot configuration, whose first robotic positioning was patented by Reymond Clavel, [6]. In Fig. 3 are presented some printers configurations. Delta configuration as many advantages when compared with the three-axis one, also called rectilinear systems. In rectilinear systems, the z movements to print each layer are produced by a platform (that moves down) or by the print head, depending on printer configuration. This means that in practice these are 2.5 axis. In Delta configuration, the platform is static and print head moves in truly 3D space, allowing incredibly much faster speeds and increased accuracy if properly constructed, [7]. Other advantages are the simpler design and easy of scaling, especially in the z direction.

Fig. 1. RepRap version I “Darwin, [5]

Fig. 2. RepRap Kossel, [8]

Basically, with projects like RepRap and many others it is possible to build a 3D printer. However producing a 3D printer that is able to produce good quality parts with a large number of materials and accurately is not so easy, especially for printers based in the delta robot. Any slight misalignment in critical components can produce large positioning errors of the print head and consequently poor quality prints. Hsieh, [1], provided some Standard Operation Procedure (SOP) to guarantee that new developed printers based in Kossel Mini project can print with and acceptable quality. This includes firmware configuration, extruder calibration and auto-bed-leveling. Alvares et. at. [9] presents a methodology to for decision-making in the design of a new linear delta robot 3D printer in order to transform requirements into engineering specifications.

84

D. Pereira et al.

Fig. 3. 3D printer types [10].

2 Printer Developed The printer presented in this work was developed based in the delta configuration. An accurately built 3D printer structure is crucial for the proper assembly of all moving parts of the printer. Furthermore, a rigorous dimensional and geometrical printer structure will save time and work on calibration procedures. The Delta IPS structure (Fig. 4) is made of V slotted reinforced commercial profiles for the columns and laser cut plates along with folded sheet parts for strengthening and components assembly. Mounting dimensions and some details about angles, offset distances, and column designation are depicted in Fig. 4. There are 3 relevant dimensional/geometrical aspects that a well-built structure must match: • All columns must be perpendicular to the top and bottom plates (detail B from Fig. 4); • Columns axes have 120° angles between them (section view A-A from Fig. 4); • All column profiles front faces have equidistant offsets from the assembled structure center axis (section view A-A from Fig. 4. To reduce at minimum mechanism clearances and backslash, all connections between rods, print head and carriage were made with magnetic balls. The carriage is built with igus® rails, whose clearance and be adjusted to control the best relation between stiffness and friction force. The print head, heated bed and heated chamber, was developed considering the process of printing high-performance materials, such as polyetheretherketone (PEEK). This kind of materials have high melt and glass transition temperatures (for polymers), and this was a design challenge taking into account the need for printing simpler materials, with significantly lower processing temperatures, like PLA (polylactic acid) or ABS (acrylonitrile butadiene styrene). The extruder is direct, water-cooled (E3D Titan Aqua), where car radiator fluid was used in a 50/50 proportion mixture with distilled water, to fill the cooling system expansion vessel until the specified level. The need for a fluid-cooled extruder is mainly related to the hot chamber temperature, which can compromise the cool zone from the hot end, melting the filament prematurely, or affect the extruder motor working temperature. Main printer specifications can be found in Table 1.

Development of a Large Size 3D Delta Printer for Advanced Polymers

85

Fig. 4. Delta IPS structure mounting dimensions and details (Dimensions in mm).

Table 1. Printer specifications Printing Area

Extrusion max. temp

Heated bed max. temp

Heated chamber Filament max. temp diameter

Max. Printing speed

400 × 850 mm2

500 °C

200 °C

80 °C

300 mm/s

1.75 mm

3 RepRap Firmware and Controller The developed 3D printer is controlled by a Duet 2 wifi printed circuit board (PCB). Duet 2 wifi is an advanced 32 bit electronic controller for 3D printers and other CNC machines [11]. This controller allows [12]: • Have a resident software driver (firmware), which is capable of transforming the Gcode commands into motor movements, control temperatures and other functional parameters; • Work as a web server, allowing calibration, upload STL files or Gcode files, control motors and temperature, directly by a computer or smartphone connected to the controller by wifi (Duet Web Control-DWC); • The printer can be controlled at a distance by configuring the router that connects it to the internet. Duet 2 wifi works with RepRap Firmware versions RRF 2.x and RRF 3.x [11]. RepRap firmware is a motion control driver that is targeted for modern 32 bit processors

86

D. Pereira et al.

and was the first open-source firmware to implement significant advances in 3D printing, including [13]: • Support for mixer extruders; • Precise timing for step pulses, even during acceleration; • Accurate extruder pressure advance, including retraction advance before the end of a move when needed; • Simulation mode, to enable accurate print times to be predicted; • Delta motion; • Least-squares auto calibration of delta printers; • Support for SPI (serial peripheral interface) controlled stepper drivers; • Heater power compensated for changes in supply voltage.

4 Calibration Printing accuracy directly depends on a proper 3D printer calibration, and it is essential to distinguish calibration procedures between two types of calibration: mechanical calibration and firmware calibration. Mechanical calibration must always be the first task to perform because if the mechanical accuracy of planes and axes is not guaranteed, the firmware calibration will not be effective as it should be. Thus, for a mechanical calibration, all parts must be accurately built and measured. A delta printer is mechanically calibrated when all the following aspects are guaranteed: • Columns are parallel to each other (maximum deference in 460 mm is 0.05 mm); • All 6 rods have the same length (maximum difference in 460 mm is 0.06 mm); • Distance between bearings on a carriage is equal to the opposed pair of bearings in the effector (maximum difference in 100 mm is 0.05 mm); • The two bearings on each carriage are at the same height; • Sliders from each carriage have the same friction coefficient; • All belts are stiff and have the same tension. In the 3D printer developed, mechanical calibration was done under an allowable dimensional deviation of 0.06mm for lengths and heights (ISO 186–1 IT7 class). In addition a precision alignment jig was built to accurately align the guides as shown in Fig. 5. The accuracy of mechanical calibration and the measures taken from the parts are the input for firmware calibration. Firmware calibration goes through the following procedures: • Adjust steps/mm value for motor drivers (maximum difference in 500 mm is 0.03 mm); • Run G32 autocalibration command, with a 4-factor scheme and 16points; • Run G29 mesh bed probe command; • Check repeatability with the G30 single z-probe command. These calibrations (firmware) were done with the DWC (duet web control). The most critical file in a board that uses RepRap firmware is the ‘config.g’ file. This file contains all information needed by the algorithm for the printer to work correctly and is where mechanical calibration and part measurements will be useful. As already referred, the

Development of a Large Size 3D Delta Printer for Advanced Polymers

87

Fig. 5. Alignment jig.

first firmware calibration step was to adjust steps/mm of the drives that control steppers motors that actuate each carriage, and this is a crucial calibration task because if the carriage displacement has a significant dimensional deviation when compared with the controller values, none of the further steps will result on accurate printing. Depicted in Fig. 6 are dimensional values that the controller needs to ‘know’ to position the effector correctly. These values are given by G32 (auto-calibration) command, and rod length (L) must not have a dimensional deviation > 0.05mm from the true length of the rods.

Fig. 6. Kinematic mechanism dimensions nomenclature

After running the G32 command several times, DWC should give the feedback, as shown in Fig. 7. Values for deviation given by DWC (Fig. 7) indicate how well the calibration procedure was, and if the printer is mechanically calibrated, values of deviation (before and after) should converge after a few G32 command runs. The 4 factors that are given by the G32 command are rod length (L), delta radius (R), endstop adjustment, and homed height. These values were obtained by simply giving both the M665 and M666 commands (without any other values), and DWC feedbacks the new values, as shown in Fig. 8.

88

D. Pereira et al.

Fig. 7. G32 command feedback values from DWC.

Fig. 8. M665 and M666 values after a G32 command run.

For bed leveling (and last calibration procedure), the G29 command runs, in fact, several G30 single z-probe commands in order to mapping bed surface heights. After G29 runs, mapped bed heights at predefined points are overwritten in the system file heighmap.csv, and when bed compensation is active, this file is where the bed compensation gets bed mapping values. In Delta IPS G29 was run with 21 points and a spacing of 60 mm between them and for each XY direction. Since bed measurements showed some discrepancies between runs, a dial gauge (with a 0.01mm resolution) (Fig. 9) was used to measure bed flatness, and it was detected that the bed glass had poor quality in terms of flatness. The discrepancy of values obtained by the G29 command between a clean and an unclean glass was analyzed, comparing the bed height mapping values obtained by the infrared sensor (IR) and values measured by the dial gauge.

Fig. 9. Dial gauge mounted on effector plate.

The plot of mean error between IR and dial gauge probing shows that the mean error and standard deviation increase with an unclean bed glass, and this fact leads to the conclusion that an IR probing sensor can induce significant errors while working with bed compensation based on IR sensor mapped values.

Development of a Large Size 3D Delta Printer for Advanced Polymers

89

Fig. 10. Mean error between IR and dial gauge measurements.

5 Z-probe Sensor Z-probe sensor has a strong influence in print quality. First, because if printer geometry corrections are needed, the printer algorithm will relay in these measurements to calculate the right dimensions for the mechanism. If these are wrong, the printer will not print accurately. Another reason is to compensate bed leveling and bed shape. In large printer like this one (400 mm diameter) and with a bed temperature of 200 ºC, the bed will deform and warp a lot. Even the initial surface is a bit warped because glass, in this case tempered glass is not plan enough. Therefore, if bed is not properly compensated first layers will not bound properly to the surface and print will fail for sure. The sensor that was first used (IR sensor) gave poor results, not enough to ensure good printing quality. It has the advantage to do measurements without contact but when the glass is not properly cleaned, the measurements are not accurate enough (as presented in Fig. 10. Another problem is because the sensor has an offset (more than 20 mm) from the nozzle. This will create an additional error if the print head does not move plan to the bed. A new contact sensor, based in strain gages is proposed to try to obtain a better precision and better bed compensation. 5.1 Infrared vs. Strain Gauge-Based Sensor In order to understand the differences between IR and strain gauge-based sensor, a brief comparison between infrared (IR) sensor [14–16], and strain gauge (SG) based sensor is made, highlighting advantages and disadvantages of each technology: IR sensor: • advantages – three different versions – i) simple threshold-based sensor; ii) differential sensor; iii) modulated IR light; – differential sensor version is not affected by interference, allowing more accurate readings against different print surfaces; – simple models are cheap; – being small, it can be directly mounted close to the hot end;

90

– – – –

D. Pereira et al.

lightweight (